ArticlePDF Available

Effects of anthropogenic salinisation on the ecological status of macroinvertebrate assemblages in the Werra River (Thuringia, Germany)

Authors:
  • German Environment Agency
  • IGF Jena, Germany, Jena

Abstract and Figures

For more than 100 years, the Werra River has been severely affected by intensive salinisation caused by potash fertilizer industries. We show considerable differences in macroinvertebrate assemblages between reaches without salinisation impact and downstream reaches with intense anthropogenic salinisation in the Werra. This is true for almost all biological metrics relevant for ecological status classification under the EU-Water Framework Directive (EU-WFD) (European Community, Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, No. L 327/1, of 22 December 2000) and diversity measures (taxon richness, evenness). Macroinvertebrate assemblages at salinisation sites were completely dominated by three halophile neobiotic macroinvertebrate species (Gammarus tigrinus, Corophium lacustre and Potamopyrgus antipodarum). We compared anthropogenically salinised sites from the Werra with disturbed but non-salinised sites from the Werra and other German rivers. We used biological metrics developed for classifying the ecological status according to the EU-WFD. This comparison indicated a severe degradation at salinisation sites on the Werra and these fell into the worst ecological status class ‘bad’ according to the EU-WFD. Multivariate statistical analyses revealed anthropogenic salinisation as a key factor causing the differences in composition of macroinvertebrate assemblages in the Werra between salinisation and reference sites. Analyses of the long-term presence–absence data of macroinvertebrate assemblages indicated no marked improvement in the ecological status in the past 20 years.
Content may be subject to copyright.
PRIMARY RESEARCH PAPER
Effects of anthropogenic salinisation on the ecological status
of macroinvertebrate assemblages in the Werra River
(Thuringia, Germany)
Jens Arle
Falko Wagner
Received: 30 January 2012 / Revised: 9 July 2012 / Accepted: 21 July 2012 / Published online: 3 August 2012
Ó Springer Science+Business Media B.V. 2012
Abstract For more than 100 years, the Werra River
has been severely affected by intensive salinisation
caused by potash fertilizer industries. We show
considerable differences in macroinvertebrate assem-
blages between reaches without salinisation impact
and downstream reaches with intense anthropogenic
salinisation in the Werra. This is true for almost all
biological metrics relevant for ecological status clas-
sification under the EU-Water Framework Directive
(EU-WFD) (European Community, Directive 2000/
60/EC of the European Parliament and of the Council
of 23 October 2000 establishing a framework for
Community action in the field of water policy, No. L
327/1, of 22 December 2000) and diversity measures
(taxon richness, evenness). Macroinvertebrate assem-
blages at salinisation sites were completely dominated
by three halophile neobiotic macroinvertebrate spe-
cies (Gammarus tigrinus, Corophium lacustre and
Potamopyrgus antipodarum). We compared anthro-
pogenically salinised sites from the Werra with
disturbed but non-salinised sites from the Werra and
other German rivers. We used biological metrics
developed for classifying the ecological status accord-
ing to the EU-WFD. This comparison indicated a
severe degradation at salinisation sites on the Werra
and these fell into the worst ecological status class
‘bad’ according to the EU-WFD. Multivariate statis-
tical analyses revealed anthropogenic salinisation as a
key factor causing the differences in composition of
macroinvertebrate assemblages in the Werra between
salinisation and reference sites. Analyses of the long-
term presence–absence data of macroinvertebrate
assemblages indicated no marked improvement in
the ecological status in the past 20 years.
Keywords Biodiversity Degradation Water
framework directive Multivariate analysis
Introduction
Salinity is a major characteristic of all waters on Earth,
which influences the living conditions of aquatic
organisms (Scho
¨
nborn, 2003). The term ‘salinity’ is
usually used for the total concentration of dissolved
inorganic ions in the water, such as Na
?
,K
?
,Mg
2?
,
Ca
2?
,Cl
-
,SO
4
2-
and HCO
3-
(Ziemann & Schulz,
2010). This total concentration is often summarized by
the electrical conductivity. The ecotoxicity of salini-
sation is usually due to the osmotic effects of the total
salt concentration and to the ion proportions
Handling editor: John M. Melack
Present Address:
J. Arle F. Wagner (&)
Institut fu
¨
r Gewa
¨
ssero
¨
kologie und Fischereibiologie Jena
(IGF), Sandweg 3, 07745 Jena, Germany
e-mail: jens.arle@uba.de
F. Wagner
e-mail: falko.wagner@igf-jena.de
123
Hydrobiologia (2013) 701:129–148
DOI 10.1007/s10750-012-1265-z
(Ziemann, 1971; Scho
¨
nborn, 2003). Although chloride
concentration is commonly applied as the unit of
measurement for salinity and the extent of salinisation
in freshwater ecosystems, Cl
-
affects the biota much
less strongly than other ions. The effects of ‘salinisa-
tion’ are the result of the combined effects of different
ions (especially potassium (K
?
), sodium (Na
?
) and
magnesium (Mg
2?
)) associated with Cl
-
or ion ratios
and the total salt concentration (Ziemann, 1997). As
they are relevant to the biological and ecological
effects of salinisation, thresholds for the ions potas-
sium (K
?
), sodium (Na
?
) and magnesium (Mg
2?
), as
well as for the total hardness, and the interactions with
temperature and pH value should be addressed in
further research. This is especially true because no
official environmental target values for these ions are
currently available in Germany or most other Euro-
pean Union member states.
The Werra has, since 1901, been affected by
salinisation caused by potash fertilizer industries
(Vogel, 1913). The river is one of the most disturbed
European freshwater ecosystems. The highest salt
loads were observed during the 1970s and 1980s. In
1976, maximum concentrations of up to 40,000 mg/l
chloride (Cl
-
) were measured (Hulsch & Veh, 1978).
Figure 1 shows the chloride concentrations (Cl
-
)
between 1979 and 2011 at two Werra sites impacted
by intensive industrial salinisation. Hu
¨
bner (2007)
reported for 1982 and 1983 maximum concentrations
of 986 mg l
-1
Mg
2?
and 584 mg l
-1
K
?
for the reach
between Witzenhausen and Hannoversch Mu
¨
nden
(Fig. 2) downstream (Letzter Heller). After the polit-
ical reunification of Germany, salt loads in the Werra
were considerably reduced. Chloride concentrations
were homogenised, with peak concentrations of
3,000 mg l
-1
(ARGE Weser, 2000). However, this
concentration still exceeds the threshold of 200
mg l
-1
Cl
-
for ‘good ecological status’ (LAWA,
2007). In the Werra, the official threshold for total
hardness has been repeatedly adjusted by regional
authorities. In 1998/1999, it was fixed at 90 German
degree (dH), which is still valid. The unit German
degree for total hardness is an expression for the sum
of the calcium and magnesium ion concentration. One
German degree of total hardness equals the concen-
tration of approximately 17.8 mg l
-1
calcium carbon-
ate (CaCO
3
). A large number of streams and rivers in
northern Thuringia are, like the Werra, affected by
anthropogenic salinisation caused by potash fertilizer
industries. This anthropogenic pressure on stream
ecosystems is thus not exclusive to the Werra. Other
freshwater ecosystems in Europe, and especially in
Australia and South Africa, also suffer from salinisa-
tion effects (Williams, 2001; Brock et al., 2005; Sim
et al., 2006; Davis et al., 2010).
Previous research has confirmed macroinverte-
brates as suitable indicators for anthropogenic salin-
isation of river ecosystems. The composition of river
communities commonly reflects the extent of salini-
sation (Short et al., 1991; Schulz, 2000; Piscart et al.,
2005, 2006; Velasco et al., 2006;Hu
¨
bner, 2007;
Braukmann & Bo
¨
hme, 2010). Thresholds derived
from controlled laboratory experiments on the salinity
tolerance of macroinvertebrates determined by LC
50
values correspond with those derived from distribution
data of field studies (Kefford et al., 2004).
We focused on quantifying the effects of intense
anthropogenic salinisation on a freshwater ecosystem.
Many other potentially relevant pressures, such as
organic and inorganic pollution, hydromorphological
degradation and salinisation, were included in the
analyses. The results from the Werra were compared
to those in other German river ecosystems with similar
common anthropogenic pressures but without inten-
sive salinisation.
We hypothesised that the impact of high mineral
salt loads in the Werra is currently the major factor
influencing the composition of macroinvertebrate
communities, causing massive degradation of the
ecological status of the original freshwater ecosystem.
Site description
The Werra is part of the Weser river catchment. The
source is located on the south side of the Thu
¨
ringer
Schiefergebirge (slate mountains) close to the Thu
¨
-
ringer Wald (Thuringian Forest) at 797 m a.s.l. After
298 km flow distance the Werra joins the river Fulda
at 117 m a.s.l. to form the river Weser. The Werra has
a catchment area of approximately 5,496 km
2
(FGG
Weser, 2005). The slope varies between 2.85% in
upstream regions, 0.245% in the middle section and
0.076% in the downstream section of the river. The
overall difference in altitude from source to mouth is
approximately 680 m. Upstream reaches of the river
are characterised by narrow, deeply cut, v-shaped
valleys with low sedimentation rates. As a result of
130 Hydrobiologia (2013) 701:129–148
123
10
100
1000
10000
100000
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2011
Time
Fig. 1 Chloride concentrations (Cl
-
) between 1979 and 2011
at two Werra sites impacted by intensive industrial salinisation
(‘Letzter Heller’—grey line (1979–1998) and ‘Gerstungen’—
black line (1993–2011). Horizontal straight lines mark: (a) the
‘official regional threshold’ for chloride at 2,500 mg l
-1
,
(b) official German recommendation value for chloride set for
the demands of the Water Framework Directive for other rivers
at 200 mg l
-1
and (c) the assumed ‘Reference conditions
(=‘Very good ecological status’) for chloride according to the
Water Framework Directive at 50 mg l
-1
. Averaged concen-
trations of chloride (2005–2008) are shown as horizontal lines
for Werra-Reference-Sites (d) and other Reference Sites from
Thuringia and Hesse (e) used in this study
Fig. 2 Map of the river
Werra, tributaries and the
sample sites used in this
study. Small red flags mark
the position of major point
releases of salt brine into the
Werra. (Color figure online)
Hydrobiologia (2013) 701:129–148 131
123
geological conditions, in particular the transition from
limestone to Bunter sandstone, the width of the valley
increases in downstream reaches, meandering in a
river bed covered with sediments of alluvial origin.
The middle and downstream sections of the Werra are
characterised by large valleys resulting from erosion
and leaching of natural mineral salts in Zechstein.
According to the stream typology of the EU-WFD
(Pottgiesser & Sommerha
¨
user, 2004; LAWA-Typol-
ogy), upstream reaches of the Werra can be classified
as a small, coarse, substrate-dominated siliceous
highland river (LAWA-Type 5). Middle sections are
classified as mid-sized, fine to coarse substrate-dom-
inated calcareous highland river sections (LAWA-
Type 9) and downstream sections can be classified as
large highland river sections (LAWA-Type 9.2).
According to the river classification on the basis of
ecological stream regions (Illies & Botosaneanu,
1963), the Werra covers zones from krenal to
epipotamal.
The potash fertilizer industries of Thuringia and
Hesse export mineral salts as point releases and in the
form of diffuse releases causing high salt loads
downstream of the town Unterrhon (Fig. 2). The
major point sources are to be found at the towns of
Dorndorf, Unterbreizbach, Hattorf and Wintershall.
Methods
We studied the Werra between kilometre 199 on the
river and Hannoversch Mu
¨
nden close to where it joins
the Fulda. All study sites and Werra reference sites
were situated in the river section classified as large
highland river (LAWA-Type 9.2)—Epipotamal.
Seven sampling sites were set within river sections
affected by anthropogenic salinisation (hereafter:
SS—Salinisation Sites), one site lay in the transitional
zone between freshwater and salinisation reaches
(hereafter: TZ—Transitional Zone), and four
upstream sites in reaches without salinisation effects
(hereafter: RS—Reference Sites). The selection of
sample sites was based on the intensity of anthropo-
genic salinisation along the river continuum as well as
the availability of physical and chemical data.
We took macroinvertebrate samples on three occa-
sions, denoted spring (i.e. 28 & 29, April 2008),
summer (16 July 2008) and autumn (1 & 2, October
2008) at 12 sampling sites (Fig. 2). Most of the
sampling sites matched sites used in the EU-WFD
monitoring programme of the regional authorities.
Thus, a broad database of physical and chemical
parameters was available for most of them. Every site
was a 20 m reach of the river. For sampling, we used
the standard methods developed for the implementa-
tion of the EU-WFD (AQEM Method, Multihabitat
Sampling) (Meier et al., 2006).
At each sampling site, the structural quality of the
reach and benthic habitat composition was assessed
using standard methods (LAWA 2002; Meier et al.,
2006). Benthic macroinvertebrates were sampled
using a square dip net (0.25 9 0.25 m with a
500 lm net, HYDRO-BIOS Apparatebau GmbH,
Kiel, Germany). Samples were preserved in 75%
ethanol, transported to the laboratory, and stored until
processing. In the laboratory, the invertebrates[2mm
were separated from the sediment–detritus mixture.
Large samples were divided into homogenous subs-
amples according to the sample processing methodol-
ogy given in the AQEM manual (AQEM consortium,
2002). The invertebrates were counted and identified to
the lowest possible taxonomic level according to the
criteria of the operational taxa list using the standard
literature (cf Haase et al., 2006a, b). At each sampling
date and site, conductivity was measured (Hanna HI
98129, Hanna Instruments, Germany).
The sampling site near Witzenhausen was originally
(first sampling in April) located immediately downstream
of the Gelster tributary, a freshwater stream without
salinisation. The April sampling results showed that a
number of freshwater species drifted into the highly
salinised Werra. Thus, the invertebrate community does
not represent the ecological status of the Werra. To avoid
this drift affecting the species data, sampling in the
summer and autumn took place 20 m upstream. The
higher variability of ecological metrics and assemblage
composition caused by this change in sampling position
at Witzenhausen did not influence the overall results and
interpretations but should be recognised.
The Thuringian regional office of environment and
geology and the Hesse regional office of environment
and geology provided additional data on macroinver-
tebrate assemblages from the EU-WFD monitoring for
Werra sites and for reference rivers of the same type
without salinisation (rivers: Saale, Weiße Elster, Un-
strut, Fulda, Eder, Schwalm and Diemel). Like our data,
these data were also based on the AQEM-method. The
data are therefore comparable between rivers.
132 Hydrobiologia (2013) 701:129–148
123
Furthermore, the regional authorities provided data
on the hydromorphological quality (structural quality
index, hereafter: SQI according to LAWA 2002), as
well as physical and chemical characteristics. The data
included information (mean, maximum, minimum,
standard deviation and number of samples) on abiotic
variables (chloride, sulphate, potassium, calcium,
magnesium, sodium, nitrate, nitrite, ammonium, ortho-
phosphate, total-phosphorous concentration, TOC and
total hardness), based on measurements made between
2005 and 2008 at intervals ranging from once a month
to once every 3 months. Data for conductivity, pH-
value, temperature and oxygen (mg l
-1
and %) were
based on data collected between 2005 and 2008.
For the assessment of the ecological status of
macroinvertebrate assemblages, we followed the
methodology developed for implementing the EU-
WFD.
The EU-WFD divides the ‘ecological status’ of
streams and rivers into five quality classes: 1 = high,
2 = good, 3 = moderate, 4 = poor and 5 = bad. The
restoration goal of the EU-WFD is to reach a ‘good
ecological status’ (class 2) in all natural streams and
rivers in Europe by 2015. The ‘ecological status’ of
macroinvertebrate assemblages was assessed using
standard approaches and standard software developed
for EU-WFD: ‘ASTERICS’ version 3.3.1. including
the module ‘PERLODES’ (Meier et al., 2006). We
used several biological metrics to describe the ‘eco-
logical status’ of macroinvertebrate assemblages. The
assessment score of the module ‘General Degradation’
is the most important metric (Arle, unpublished data)
within the multimetric approach to determine the
‘overall ecological status’ in line with EU-WFD
requirements in Germany.
Taxa density (number of taxa per sample), the
Simpson’s index of diversity (1-D), the Shannon-
Wiener-Index, the accumulated number of taxa (based
on pooled seasonal data sets) and the ‘expected taxon
richness’ were used as variables to characterise
invertebrate diversity. Because the species and taxon
richness measured depends partly on sampling effort
(Gotelli & Colwell, 2001), we used Ecosim simulation
software (Gotelli & Entsminger, 2001) to estimate the
‘expected taxon richness’ for a given number of
individuals drawn randomly from a sample (McCabe
& Gotelli, 2000). Ecosim performs a Monte Carlo
method similar to rarefaction (Hurlbert, 1971; Sim-
berloff, 1972). The smallest sample (based on pooled
seasonal samples) consisted of 2,940 individuals,
therefore 2,940 individuals were randomly sampled
from each sample, and the number of taxa observed
was recorded. The randomisation was repeated 100
times for each sample, and the average number of taxa
was used as the ‘expected taxon richness’.
The data was statistically analysed using the
software package Sigma Stat 2.0 (SPSS Cooperation).
Normal distribution of the data was tested by
Kolmogorov–Smirnov tests and equivalency of vari-
ances using Levene tests. Analysis of Variance
(ANOVA) and Tukey’s Post hoc tests (P \ 0.05) or
Kruskal–Wallis test followed by Dunn’s Tests
(P \ 0.05) were used to compare variables between
groups of sites. Pearson correlations were used to
reveal relationships of invertebrate data (mean scores
of the module ‘General Degradation’ and mean taxon
number with abiotic variables at a level of P \ 0.05).
The macroinvertebrate assemblage and environmental
factor data for five additional Werra sampling sites
were included in correlation and ordination analyses.
These were sampling sites which were part of the
official EU-WFD monitoring, comprised three refer-
ence sampling sites without salinisation (Wasungen,
Meiningen, and Dorndorf) and two sites severely
affected by salinisation (Dorndorf & Heringen)
(Fig. 2).
Ordination statistics were calculated with the
program Canoco Version 4 (ter Braak & Smilauer,
1998). To assess variation in the biological metric
composition (for definitions of the metrics used, see
RDA description) between sample sites, a principal
component analysis (PCA) (Jongman et al., 1995) was
conducted. In order to estimate the relative importance
of environmental variables for the composition of
biological metrics of the invertebrate assemblages and
consequently, the ecological status (metrics of the
ecological status classification according to EU-
WFD), multivariate Redundancy Analyses (RDA)
and Monte Carlo permutation procedures (Jongman
et al., 1995) were used. The following biological
metrics were included in the analysis: Abundance,
Number of Taxa, German Saprobic Index (new
version), Diversity (Simpson-Index), Potamon Type
Index with abundance classes, r/K relationship, meta-
rhithral, metarhithral (scored taxa = 100%), Rheoin-
dex (Banning, with abundance), number of indicator
taxa salinity preference, freshwater, oligohaline, mes-
ohaline, polyhaline, euhaline, EPT-Taxa, EPT
Hydrobiologia (2013) 701:129–148 133
123
(abundance classes), EPTCBO (Eph., Ple., Tri., Col.,
Bivalv., Odo.), Score General Degradation, Score
German Fauna Index type 9.2, Score metarhithral
(scored taxa = 100%), Score EPT abundance classes)
and Score EPTCBO (Eph., Ple., Tri., Col., Bivalv.,
Odo.) (for an explanation of these metrics see Meier
et al., 2006). A preliminary analysis (Detrended
Correspondence Analysis—DCA) showed that varia-
tion in macroinvertebrate assemblage data was best
described by linear rather than by unimodal models
(length of gradients was 1.446). Thus, a RDA com-
bined with a stepwise forward selection with Monte
Carlo permutations (199), including all environmental
variables, was carried out to extract a reduced variable
set, to select variables that explained most of the
variability in composition of biological metrics of the
invertebrate assemblages, and to test for significance
of environmental variables. Forward selection on
all environmental factors showed that all factors
associated with anthropogenic salinisation explained
a major part of the variability. A second RDA was
conducted, including environmental factors with sig-
nificant ‘conditional effects’ determined by forward
selection.
A second PCA was conducted on the presence
absence of taxa in our datasets and historical ones
(Meinel & Barlas, 1987; Fesel et al., 1996) in order to
investigate the long-term changes in macroinverte-
brate community composition.
Results
All environmental factors associated with anthropo-
genic salinisation changed rapidly downstream of the
first operational discharge near Vacha (Table 1). Con-
centrations of chloride, potassium, sodium, magnesium
and sulphate, as well as conductivity and total hardness,
increased substantially in comparison to unaffected
reference sites. Only calcium concentration increased to
a lesser extent. With the exception of TOC concentra-
tions, which also increased slightly downstream, there
were no visible trends along the river in the environ-
mental factors (NH
4
?
,BSB
5,
NO
2
-
,NO
3
-
,o-PO
4
-
,
P
total
, pH and oxygen) not directly associated with
anthropogenic salinisation (Table 1). The hydromor-
phological quality scores (SQI) varied between 2 (near
natural) at Barchfeld and Gerstungen, and 7 (exces-
sively disturbed) at Lauchro
¨
den. The composition of
mesohabitats (Choriotope type) varied widely without a
clear downstream pattern.
In total, 121 invertebrate taxa were found during the
study. The number of taxa at SS was considerably lower
in comparison to RS sites (Fig. 3). At the TZ (Vacha)
between freshwater RS upstream and SS downstream,
an intermediate number of taxa was found.
At RS the mean number of taxa (per sample) ranged
between 30 and 40, while the mean number of taxa at
SS was considerably lower (Fig. 3). Accumulated
numbers of taxa based on pooled samples and
estimated taxa numbers were considerably higher at
upstream RS. The macroinvertebrate communities at
SS were dominated by three halophile and neobiotic
species: Gammarus tigrinus (Sexton, 1939), Coroph-
ium lacustre (Vanho
¨
ffen, 1911) and Potamopyrgus
antipodarum (Gray, 1843). Freshwater taxa accounted
for lower proportions of the assemblages than meso-,
poly- and euhaline taxa, with only single individuals
or very low numbers of the freshwater taxa per site. At
all upstream freshwater sites (RS), ASTERICS indi-
cated a good ecological status with low seasonal
variability (Table 2). Scores of the module ‘General
Degradation’, the German Fauna Index for River
Type 9.2, the proportion of taxa with preferences for
metarhithral river regions ([%] metarhithral (scored
taxa = 100%)), the proportion of Ephemeroptera,
Plecoptera and Trichoptera taxa (EPT [%] (abundance
classes), diversity (Simpson-Index, Shannon-Wiener-
Index, Number of Taxa, Accumulated Number of
Taxa, Estimated Number of Taxa) and the proportion
of freshwater and oligohaline taxa were considerably
higher at all freshwater RS than in SS. German
Saprobic Index and Abundance (Ind m
-2
) were lower
and the proportion of meso-, poly- and euhaline taxa
(%) was almost zero at freshwater RS. However, it
must be noted that the saprobic index at SS was mainly
influenced by a single indicator taxon (Gammarus
tigrinus). Because of the dominance of this taxon at
salinisation sites, the saprobic index calculated is not
an appropriate indicator for organic pollution at these
sites. Higher proportions of freshwater and oligohaline
macroinvertebrate taxa at Witzenhausen were caused
by the invertebrate drift from the Gelster tributary in
April (see Methods).
Comparison of freshwater sites (RS—Werra,
n = 20 and Reference sites—from other rivers in
Thuringia and Hesse, n = 56) with SS from the Werra
(n = 23) revealed that all freshwater sites in Thuringia
134 Hydrobiologia (2013) 701:129–148
123
Table 1 Environmental characteristics of the sample sites
Sample site
Wernshausen Breitungen Barchfeld Tiefenort Vacha Gerstungen
Affected by salinisation No No No No Yes Yes
Northing
a
5,621,369 5,625,900 5,629,920 5,633,789 5,633,700 5,649,608
Easting
a
3,596,159 3,593,660 3,591,290 3,582,089 3,573,879 3,576,889
Stream kilometre 200 194 189 175 163 133
Structural Quality Index—
SQI
352542
Slope (%) 0.593 0.171 0.054 0.182 0.11 0.21
Height a.s.l. (m) 246 245 242 234 224 206
Choriotop—type (%)
Megalithal 1.67 1.67 6.67 16.67 1.67 1.67
Macrolithal 10.00 13.33 26.67 31.67 13.33 36.67
Mesolithal 51.67 46.67 8.33 13.33 15.00 25.00
Microlithal 16.67 18.33 8.33 16.67 21.67 15.00
Akal 11.67 13.33 11.67 8.33 20.00 8.33
Psammal/Psammopelal 3.33 5.00 35.00 1.66 16.67 8.33
Argyllal 1.67
Technolithal 6.67 1.67
Others 5.00 1.67 3.33 5.00 10.00 3.33
Conductivity (lScm
-1
) 526 ± 122 525 ± 120 506 ± 120 637 ± 172 2,320 ± 717 6,160 ± 1,200
BSB
5
(mg l
-1
) 2.16 ± 1.13 2.47 ± 1.17 2.74 ± 1.22 3.98 ± 2.82 2.00 ± 0.27 1.81 ± 1.11
Ca
2?
(mg l
-1
) 64.40 ± 13.30 63.40 ± 14.00 63.70 ± 14.90 63.80 ± 13.70 77.60 ± 20.40 90.30 ± 51.00
Cl
-
(mg l
-1
) 40.0 ± 15.1 39.0 ± 13.9 39.2 ± 13.1 74.8 ± 29.6 597.0 ± 234.0 1680.0 ± 508.0
SO
4
2-
(mg l
-1
) 61.2 ± 18.0 62.8 ± 19.2 64.5 ± 20.2 64.7 ± 18.9 114.0 ± 39.7 333.0 ± 86.4
Total hardness (°dH) 11.60 ± 2.56 11.60 ± 2.67 11.50 ± 2.84 11.60 ± 2.61 39.80 ± 12.80 67.60 ± 13.70
K
?
(mg l
-1
) 3.68 ± 0.94 3.63 ± 0.86 3.62 ± 0.94 4.08 ± 1.21 36.5 ± 16.3 141.0 ± 1.88
Mg
2?
(mg l
-1
) 11.00 ± 2.72 11.30 ± 2.88 11.10 ± 3.09 11.60 ± 2.96 126.0 ± 57.90 238.0 ± 60.0
Na
?
(mg l
-1
) 20.9 ± 8.39 20.9 ± 7.96 19.7 ± 8.01 43.7 ± 24.9 168.0 ± 109.0 742.0 ± 7.82
NH
4
?
(mg l
-1
) 0.18 ± 0.08 0.17 ± 0.07 0.22 ± 0.14 0.20 ± 0.11 0.22 ± 0.12 0.22 ± 0.12
NO
2
-
(mg l
-1
) 0.046 ± 0.021 0.049 ± 0.020 0.051 ± 0.019 0.051 ± 0.021 0.051 ± 0.019 0.053 ± 0.018
NO
3
-
(mg l
-1
) 3.33 ± 0.39 3.26 ± 0.44 3.37 ± 0.41 3.39 ± 0.42 3.40 ± 0.55 3.37 ± 1.07
O
2
(mg l
-1
) 10.6 ± 1.61 10.2 ± 1.66 10.7 ± 1.55 10.4 ± 1.89 10.9 ± 1.86 10.2 ± 2.13
o-PO
4
-
(mg l
-1
) 0.136 ± 0.061 0.134 ± 0.066 0.141 ± 0.070 0.143 ± 0.073 0.142 ± 0.074 0.149 ± 0.061
P
total
(mg l
-1
) 0.182 ± 0.062 0.173 ± 0.069 0.186 ± 0.069 0.184 ± 0.079 0.176 ± 0.075 0.204 ± 0.064
pH 7.84 ± 0.30 7.85 ± 0.25 7.90 ± 0.22 7.85 ± 0.24 7.89 ± 0.23 7.85 ± 0.20
TOC (mg l
-1
) 4.43 ± 1.95 4.04 ± 1.31 3.86 ± 1.49 4.14 ± 1.28 4.32 ± 1.62 5.40 ± 2.40
Sample site
Lauchro
¨
den Frankenroda Frieda Eschwege
Niederohne
Witzenhausen Hannoversch
Mu
¨
nden
Affected by salinisation Yes Yes Yes Yes Yes Yes
Northing
a
5,651,513 5,663,299 5,673,179 5,675,650 5,690,522 5,697,255
Easting
a
3,580,583 3,589,339 3,579,050 3,570,224 3,560,068 3,548,315
Stream kilometre 122 89 64 52 20 4
Structural Quality Index—SQI 7 5 6 6 6 5
Hydrobiologia (2013) 701:129–148 135
123
and Hesse without strong salinisation, despite other
common pressures such as organic pollution or
structural degradation, showed on average higher
scores in the module ‘General Degradation’, a higher
German Fauna Index for River-Type 9.2 and
larger proportions of taxa with preferences for meta-
rhithral river regions ([%] metarhithral (scored
taxa = 100%)). They also had a larger proportion of
Ephemeroptera, Plecoptera and Trichoptera taxa (EPT
[%] (abundance classes)), higher diversity (Number of
Taxa, Higher Number of Taxa of Ephemeroptera,
Plecoptera, Trichoptera, Coleoptera, Bivalvia and
Odonata-EPTCBO (Eph., Ple., Tri., Col., Bivalv.,
Odo.)), lower saprobity (German Saprobic Index-new
version) and lower abundance (ind m
-2
) than SS
(Table 3).
Table 1 continued
Sample site
Lauchro
¨
den Frankenroda Frieda Eschwege
Niederohne
Witzenhausen Hannoversch
Mu
¨
nden
Slope (%) 0.158 1.1 0.059 n.a. 0.252 0.422
Height a.s.l. (m) 191 187 165 156 150 131
Choriotop—type (%)
Megalithal 63.33 1.67 1.67 1.67 5.00
Macrolithal 15.00 10.00 6.67 20.00 13.33 43.33
Mesolithal 8.33 21.67 31.67 38.33 30.00 15.00
Microlithal 8.33 10.00 30.00 13.33
Akal 1.67 8.33 6.67 5.00 5.00 1.67
Psammal/Psammopelal 3.33 43.33 21.67 20.00 48.33 21.67
Argyllal
Technolithal 8.33
Others 5.00 1.67 1.67 3.33 5.00
Conductivity (lScm
-1
) 6,260
b
5,220 ± 1,600 4,628
b
3,750
b
4,386
b
4,197
b
BSB
5
(mg l
-1
) n.a 2.63 ± 1.34 n.a n.a 2.83 ± 0.35 n.a
Ca
2?
(mg l
-1
) n.a 107.00 ± 31.00 n.a n.a 118.32 ± 28.45 n.a
Cl
-
(mg l
-1
) n.a 1500.0 ± 445.0 n.a n.a 1155.5 ± 165.1 n.a
SO
4
2-
(mg l
-1
) n.a 343.0 ± 108.0 n.a n.a 292.5 ± 45.2 n.a
Total hardness (°dH) n.a 63.40 ± 9.46 n.a n.a 53.44 ± 4.74 n.a
K
?
(mg l
-1
) n.a 136.0 ± 43 n.a n.a 91.3 ± 18.6 n.a
Mg
2?
(mg l
-1
) n.a 210.0 ± 40.9 n.a n.a 161.0 ± 27.0 n.a
Na
?
(mg l
-1
) n.a 601.0 ± 185.0 n.a n.a 528.7 ± 106.1 n.a
NH
4
?
(mg l
-1
) n.a 0.18 ± 0.12 n.a n.a 0.10 ± 0.10 n.a
NO
2
-
(mg l
-1
) n.a 0.045 ± 0.028 n.a n.a n.a. n.a
NO
3
-
(mg l
-1
) n.a 3.49 ± 0.88 n.a n.a 3.40 ± 0.80 n.a
O
2
(mg l
-1
) n.a 11.4 ± 1.48 n.a n.a 9.90 ± 1.80 n.a
o-PO
4
-
(mg l
-1
) n.a 0.132 ± 0.057 n.a n.a 0.200 ± 0.100 n.a
P
total
(mg l
-1
) n.a 0.177 ± 0.064 n.a n.a 0.300 ± 0.200 n.a
pH n.a 8.06 ± 0.33 n.a n.a 8.30 ± 0.30 n.a
TOC (mg l
-1
) n.a 5.25 ± 1.71 n.a n.a 5.70 ± 2.20 n.a
Physical and chemical characteristics of the water are given as mean ± one standard deviation. n.a. not available
a
Coordinates given for Germany Zone 3, Projection: Transverse Mercator, Datum: Deutsches Hauptdreiecksnetz
b
Measured once in June 2008
136 Hydrobiologia (2013) 701:129–148
123
A comparison of abiotic variables among freshwa-
ter sites (RS—Werra and Reference sites—from other
rivers in Thuringia and Hesse) and salinisation sites
(SS) from the Werra revealed that salinisation asso-
ciated variables alone, and no other measured envi-
ronmental variables or stressors were able to explain
the differences in macroinvertebrate community
observed among the three groups of sites (Werra
references sites, Reference sites from other rivers and
salinisation sites (SS)) (Table 4). There were no
statistically significant differences in hydromorpho-
logical degradation (SQI—index and habitat compo-
sition) or organic pollution between reference sites
and salinisation sites (Table 4).
Biological metrics correlated significantly with all
salinisation-associated factors (Table 5) along the
Werra. In contrast, other environmental variables
showed low or no significant correlations with the
module ‘General Degradation’ or the number of taxa.
The first two PCA axes explained 90% of the
variability in the biological metric composition of
the invertebrate assemblages. The first axis (85%)
represented the salinisation gradient as indicated by
the visible separation of RS and SS. The biological
metric composition at the TZ (Vacha) showed greater
similarities to those of RS (Fig. 4).
In the RDA, including all environmental vari-
ables, the first two axes explained 76% of the
variability in the data on the composition of
biological metrics of the invertebrate assemblages,
and the first axis contributed the majority (73%)
(Table 6). The relationship between biological met-
ric data of the invertebrate assemblages and explan-
atory environmental variables accounted for 90%
(axis 1), indicating a strong correlation between the
scores of environmental variables and biological
metric data. After stepwise forward selection, only
the maximum potassium concentration (K
max
?
) and
the standard deviation of total hardness (°dH Std.)
still had significant effects on the composition of
biological metrics, with the former being the most
important. The maximum potassium concentration
was the variable that explained the largest percent-
age of variability (71%) in the composition of
Sampling site
W
e
r
n
s
h
a
u
s
e
n
u
h
B
r
e
i
t
u
n
g
e
n
B
a
r
c
h
f
e
l
d
Ti
e
f
e
n
o
r
t
V
a
c
h
a
G
e
r
s
t
u
n
g
e
n
L
a
u
c
h
r
ö
d
e
n
F
r
a
n
k
e
n
r
o
d
a
F
r
i
e
d
a
E
s
c
h
w
e
g
e
N
i
e
d
e
r
o
h
n
e
W
i
t
z
e
n
h
a
u
s
e
n
H
a
n
n
M
ü
n
d
e
n
Number of taxa
0
20
40
60
80
100
pooled data
mean (n=3)
ecosim (2940 Ind.)
Reference sites,
no salinisation,upstream
Salinisation sites,
downstream
Transitional
zone
Fig. 3 Number of taxa
found at 12 sampling sites in
2008 (open circles: mean
number of taxa per 1.25 m
2
;
filled circles: accumulated
number of taxa based on
pooled data, and filled
triangles: estimated number
of taxa per 2,940 ind. based
on pooled data)
Hydrobiologia (2013) 701:129–148 137
123
Table 2 Biological characteristics (mean ± one standard deviation) of the sample sites
Variable Sample site
Wernshausen Breitungen Barchfeld Tiefenort Vacha Gerstungen
Ecological status class Good Good Good Good Moderate Bad
General degradation score 0.82 ± 0.08 0.80 ± 0.04 0.53 ± 0.20 0.70 ± 0.06 0.50 ± 0.07 0.00 ± 0.00
German Fauna Index type 9.2 0.79 ± 0.12 0.85 ± 0.12 0.64 ± 0.25 0.77 ± 0.06 0.53 ± 0.15 0.00 ± 0.00
Metarhithral (%) (scored taxa = 100%) 0.82 ± 0.16 0.76 ± 0.08 0.75 ± 0.27 0.62 ± 0.23 0.32 ± 0.04 0.00 ± 0.00
EPT (%) (abundance classes) 0.73 ± 0.24 0.67 ± 0.25 0.14 ± 0.05 0.52 ± 0.22 0.31 ± 0.12 0.00 ± 0.00
EPTCBO (Eph., Ple., Tri., Col., Bivalv., Odo.) 0.98 ± 0.03 0.82 ± 0.28 0.37 ± 0.14 0.73 ± 0.20 0.77 ± 0.06 0.00 ± 0.00
German Saprobic Index (new version) 1.91 ± 0.04 1.94 ± 0.03 2.06 ± 0.03 1.98 ± 0.01 2.02 ± 0.03 2.27 ± 0.10
Diversity (Simpson-Index) 0.84 ± 0.09 0.82 ± 0.09 0.77 ± 0.05 0.83 ± 0.07 0.80 ± 0.04 0.33 ± 0.36
Diversity (Shannon-Wiener-Index) 2.56
± 0.61 2.36 ± 0.31 2.01 ± 0.37 2.26 ± 0.35 2.12 ± 0.20 0.63 ± 0.72
Abundance (ind. 1.25 m
-2
) 965 ± 762 2,942 ± 3,500 1,076 ± 300 961 ± 756 1,840 ± 1,175 6,243 ± 5,383
Number of Taxa 47.33 ± 10.69 36.33 ± 6.43 27.00 ± 6.56 31.33 ± 7.02 34.33 ± 2.08 9.00 ± 2.08
Accumulated number of taxa based on pooled data 87 67 51 55 62 21
Estimated number of taxa * (2,940 Ind. using Ecosim) 86.92 ± 0.28 55.00 ± 2.23 50.91 ± 0.32 56.00 ± 0.00 56.42 ± 1.84 10.33 ± 1.76
Proportion of freshwater & oligohaline taxa (%) 98.5 ± 2.44 99.99 ± 0.02 99.91 ± 0.15 96.94 ± 4.76 87.32 ± 3.95 42.63 ± 2.94
Proportion of meso, poly- & euhaline taxa (%) 1.49 ± 2.44 0.01 ± 0.02 0.09 ± 0.15 3.06 ± 4.76 12.57 ± 3.77 49.84 ± 12.68
Variable Sample site
Lauchro
¨
den Frankenroda Frieda Eschwege Niederohne Witzenhausen Hannoversch
Mu
¨
nden
Ecological status class Bad Bad Bad Bad Bad Bad
General degradation score 0.00 ± 0.00 0.01 ± 0.01 0.08 ± 0.07 0.03 ± 0.03 0.09 ± 0.16 0.01 ± 0.01
German Fauna Index type 9.2 0.00 ± 0.00 0.00 ± 0.00 0.06 ± 0.10 0.00 ± 0.00 0.10 ± 0.17 0.00 ± 0.00
Metarhithral (%) (scored taxa =
100%) 0.00 ± 0.00 0.05 ± 0.09 0.02 ± 0.03 0.00 ± 0.00 0.12 ± 0.20 0.04 ± 0.07
EPT (%) (abundance classes) 0.00 ± 0.00 0.00 ± 0.00 0.26 ± 0.44 0.16 ± 0.12 0.00 ± 0.00 0.00 ± 0.00
EPTCBO (Eph., Ple., Tri., Col., Bivalv., Odo.) 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.03 ± 0.06 0.13 ± 0.23 0.00 ± 0.00
German Saprobic Index (new version) 2.35 ± 0.03 2.26 ± 0.08 2.16 ± 0.09 2.21 ± 0.07 2.19 ± 0.26 2.34 ± 0.06
Diversity (Simpson-Index) 0.15 ± 0.22 0.33 ± 0.40 0.16 ± 0.22 0.17 ± 0.08 0.63 ± 0.18 0.50 ± 0.28
Diversity (Shannon-Wiener-Index) 0.30 ± 0.39 0.68 ± 0.83 0.41 ± 0.55 0.39 ± 0.15 1.29 ± 0.71 0.97 ± 0.67
Abundance (ind. 1.25 m
-2
) 6,097 ± 6,511 13,115 ± 12,029 22,041 ± 28,825 12,985 ± 11,403 4,744 ± 5,807 2,936 ± 3825
Number of Taxa 3.33 ± 0.58 5.33 ± 3.21 7.00 ± 4.36 9.67 ± 2.08 11.00 ± 9.54 6.67 ± 3.21
Accumulated number of taxa based
on pooled data
612 15 23 2516
138 Hydrobiologia (2013) 701:129–148
123
biological metrics and showed the highest correlation
(0.942) with the first canonical factor. Other vari-
ables indicating additional anthropogenic impacts,
such as nutrients (N and P), organic pollution (TOC
and BSB
5
) or hydromorphological degradation
(structural quality index), showed considerably lower
correlations (Lambda 1, Table 7). In the second
RDA, including only the two environmental vari-
ables determined by forward selection, the first two
axes explained 73% of the variability in data on the
composition of biological metrics of the invertebrate
assemblages, and the first axis contributed 72%.
Monte Carlo permutations tests showed that the
ordination axes of the RDA were significant
(Table 8). The relationships described above are
visualised in a triplot (Fig. 5). The RDA showed that
maximum potassium concentration explained most
of the variability in the composition of biological
metrics of invertebrate assemblages.
The extremely high covariance of the factor
K
max
?
with all other associated ion concentrations and
variables (Na
?
,K
?
,Mg
2?
,Ca
2?
,Cl
-
,SO
4
2-
, total
hardness and conductivity) observed in the first RDA
identifies this factor as a surrogate for salinisation in
general.
This high covariance is a result of the chemical
composition of the salt brine released. Most of these
salinisation-associated factors can be substituted, leading
only to low or marginal changes in the total variability
explained by the ordination analysis. This was especially
true for the mean potassium concentration (K
mean
?
)and
the mean chloride concentration (Cl
mean
-
), which showed
slightly lower ‘Marginal Effects’ in comparison to K
max
?
,
indicating that slightly less variance was explained by
Cl
mean
-
(lamba1 = 0.69) and K
mean
?
(lamba1 = 0.70) in
comparison to K
max
?
(lamba1 = 0.71) when the particular
variable was used as the only environmental variable
(Table 7).
In the ordination analysis of long-term changes in
macroinvertebrate community composition based on
the presence–absence data of taxa, the first two axes
explained 52% of the variability in composition of
assemblages (Fig. 6). The first axis contributed the
majority (44%), representing the salinisation gradient
as indicated by the visible separation of RS and SS.
The slight shifts of the data points from the Werra in
1985, 1991, 1995 and the recent SS data (Fig. 6)
indicated that species composition has not changed
markedly during the last 20 years.
Table 2 continued
Variable Sample site
Lauchro
¨
den Frankenroda Frieda Eschwege Niederohne Witzenhausen Hannoversch
Mu
¨
nden
Estimated number of taxa * (2,940 Ind.
using Ecosim)
5.54 ± 1.09 4.26 ± 0.57 7.33 ± 1.23 13.34 ± 1.46 15.75 ± 1.71 12.48 ± 1.43
Proportion of freshwater &
oligohaline taxa (%)
40.30 ± 0.27 47.35 ± 9.24 40.72 ± 0.29 41.54 ± 1.45 66.56 ± 22.08 49.12 ± 13.20
Proportion of meso, poly- &
euhaline taxa (%)
56.41 ± 5.35 49.19 ± 15.21 56.20 ± 5.38 56.34 ± 1.76 25.15 ± 11.15 43.14 ± 15.03
Hydrobiologia (2013) 701:129–148 139
123
Table 3 Comparison of biological characteristics between reference sites and salinisation sites
Variable Reference sites
a
(other rivers
in Thuringia & Hesse, n = 56)
Reference sites*
(Werra, n = 20)
Salinisation sites
(Werra, n = 23)
General Degradation score 0.48 ± 0.24
B
0.73 ± 0.13
A
0.04 ± 0.07
C
German Fauna Index type 9.2 0.43 ± 0.30
B
0.74 ± 0.13
A
0.02 ± 0.07
C
(%) metarhithral (scored taxa = 100%) 0.53 ± 0.27
B
0.79 ± 0.20
A
0.04 ± 0.08
C
EPT (%) (abundance classes) 0.41 ± 0.31
B
0.69 ± 0.32
A
0.09 ± 0.25
C
EPTCBO (Eph., Ple., Tri., Col., Bivalv., Odo.) 0.61 ± 0.25
A
0.70 ± 0.23
A
0.02 ± 0.09
B
German Saprobic Index (new version) 2.07 ± 0.11
B
1.98 ± 0.06
C
2.25 ± 0.13
A
Abundance (Ind. 1.25 m
-2
) 1283.82 ± 1338.21
B
1240.60 ± 1405.23
B
8976.35 ± 12432.58
A
Number of Taxa 32.02 ± 8.08
A
31.80 ± 9.37
A
7.35 ± 4.33
B
Different capital letters indicate significant differences (ANOVA; P \ 0.001 and Tukey’s Posthoc test P \ 0.05)
a
Without industrial salinisation, but with other anthropogenic pressures
Table 4 Comparison of physical and chemical characteristics (mean ± one standard deviation; median and number of sample sites)
between reference sites and salinisation sites
Variable Reference sites (other rivers
in Thuringian & Hesse)
Reference sites (Werra) Salinisation sites (Werra)
Structural Quality
Index-SQI
5.29 ± 1.49; 5.00 (n = 14) 4.43 ± 1.14; 5.00 (n = 7) 5.00 ± 1.50; 5.00 (n = 10)
BSB
5
(mg l
-1
) 2.82 ± 0.37; 2.82 (n = 14) 3.00 ± 0.60; 3.20 (n = 7) 2.47 ± 0.76; 2.32 (n = 6)
Ca
2?
(mg l
-1
) 88.01 ± 67.34; 48.41 (n = 14) 61.23 ± 4.77; 63.70 (n = 7) 90.71 ± 18.69; 86.18 (n = 6)
Cl
-
(mg l
-1
) 112.90 ± 115.71; 65.67
(n = 14)
B
39.74 ± 16.53; 39.00
(n = 7)
B
1198.59 ± 450.94; 1327.77
(n = 6)
A
SO
4
2-
(mg l
-1
) 158.87 ± 153.87; 82.98
(n = 14)
AB
58.20 ± 7.16; 61.20
(n = 7)
A
251.74 ± 108.82; 304.21 (n = 6)
B
Total hardness (°dH) 17.45 ± 12.88; 11.20 (n = 14)
B
11.04 ± 0.95; 11.50
(n = 7)
B
55.10 ± 11.36; 58.01 (n = 6)
A
K
?
(mg l
-1
) 7.26 ± 5.22; 5.04 (n = 14)
B
3.66 ± 0.22; 3.63 (n = 7)
B
98.33 ± 51.91; 113.64 (n = 6)
A
Mg
2?
(mg l
-1
) 22.34 ± 15.33; 14.98 (n = 14)
B
10.56 ± 1.22; 11.10
(n = 7)
B
184.66 ± 45.41; 185.48 (n = 6)
A
Na
?
(mg l
-1
) 59.08 ± 48.43; 49.45 (n = 14)
B
21.46 ± 10.20; 19.70
(n = 7)
B
395.95 ± 258.96; 348.34 (n = 6)
A
NH
4?
(mg l
-1
) 0.12 ± 0.09; 0.08 (n = 14)
A
0.21 ± 0.06; 0.20 (n = 7)
B
0.18 ± 0.03; 0.18 (n = 6)
AB
NO
3-
(mg l
-1
) 5.24 ± 4.16; 4.28 (n = 14)
A
3.21 ± 0.21; 3.27 (n = 7)
B
3.40 ± 0.04; 3.39 (n = 6)
AB
O
2
(mg l
-1
) 10.49 ± 0.52; 10.35 (n = 14) 10.54 ± 0.20; 10.60 (n = 7) 10.44 ± 0.59; 10.25 (n = 6)
P
total
(mg l
-1
) 0.19 ± 0.07; 0.17 (n = 14) 0.18 ± 0.01; 0.18 (n = 7) 0.23 ± 0.06; 0.19 (n = 6)
pH 7.83 ± 0.15; 7.77 (n = 14) 7.84 ± 0.07; 7.85 (n = 7) 7.93 ± 0.19; 7.87 (n = 6)
TOC (mg l
-1
) 5.18 ± 0.74; 5.30 (n = 14)
A
4.03 ± 0.25; 4.04 (n = 7)
B
4.75 ± 0.84; 4.80 (n = 6)
AB
Megalithal (%) n.a. 6.67 ± 7.07; 4,17 (n = 4) 9.58 ± 21.76; 1.67 (n = 8)
Macrolithal (%) n.a. 20.42 ± 10.40; 20.00(n = 4) 19.79 ± 13.17; 14.17 (n = 8)
Mesolithal (%) n.a. 30.00 ± 22.32; 30.00
(n = 4)
23.13 ± 10.06; 23.33 (n = 8)
Microlithal (%) n.a. 15.00 ± 5.51; 16.67 (n = 4) 12.29 ± 10.23; 11.67 (n = 8)
140 Hydrobiologia (2013) 701:129–148
123
Discussion
There were considerable differences in macroinverte-
brate assemblages between reaches with and those
without intensive anthropogenic salinisation. This was
true for the ecological status classification under the
EU-WFD as well as for diversity measures (taxon
richness and evenness). Macroinvertebrate assem-
blages at salinisation sites were dominated by three
abundant halophilic neozoic macroinvertebrate spe-
cies G. tigrinus, C. lacustre and P. antipodarum.
Gammarus tigrinus is one of the few organisms that
are able to regulate ion concentrations (especially
potassium) and are therefore adapted to variable ion
concentrations (Koop, 1991, 1994). This species was
introduced into the Werra in 1957 (Schmitz, 1960)to
compensate for losses in ecological functioning in the
heavily degraded river. The presence of C. lacustre
was reported for the first time in 1993 by Ba
¨
the et al.
(1994) and since then this species has continued to
spread upstream in the salinised part of the Werra.
Potamopyrgus antipodarum was first reported in 1949
(Fittkau, 1950). This species is currently abundant in
the salinised parts of the Werra but rare in the
Table 5 Pearson
correlation coefficients
between average biological
metrics and environmental
variables (P \ 0.05,
n = 13)
For all chemical variables,
described by mean,
maximum and minimum,
only the variable with the
highest correlation
coefficient is shown. The
three highest Pearson
correlation coefficients are
italicised. n.s. not
significant
Environmental variable ‘General degradation’ Taxon number
K
mean
-0.961 -0.906
SO
4mean
-0.978 -0.905
Mg
min
-0.957 -0.871
CL
mean
-0.948 -0.889
°dH
min
-0.946 -0.848
Cond.
min
-0.939 -0.911
Ca
max
-0.900 -0.737
Ca/Mg-ratio
min
0.873 0.741
Na
max
-0.865 -0.793
NO
3max
-0.787 -0.741
TP
mean
-0.642 -0.555
TOC
mean
-0.578 n.s.
NH
4min
n.s. 0.600
o-PO
4min
n.s. -0.604
Structural Quality Index (SQI) n.s. n.s.
BSB 5
mean, max & min
n.s. n.s.
NO
2mean, max & min
n.s. n.s.
O
2
-PN
mean, max & min
n.s. n.s.
NO
2mean, max & min
n.s. n.s.
pH
mean, max & min
n.s. n.s.
K/Na-ratio
mean, max & min
n.s. n.s.
Table 4 continued
Variable Reference sites (other rivers
in Thuringian & Hesse)
Reference sites (Werra) Salinisation sites (Werra)
Akal (%) n.a. 11.25 ± 2.10; 11.67 (n = 4) 7.08 ± 5.82; 5.83 (n = 8)
Psammal/Psammopelal
(%)
n.a. 11.25 ± 15.89; 4.17 (n = 4) 22.92 ± 15.63, 20.83 (n = 8)
Other substrates (%) n.a. 5.42 ± 4.38; 4.17 (n = 4) 5.21 ± 4.83; 4,17 (n = 8)
Different capital letters indicate significant differences (Kruskal–Wallis Test; P \ 0.01 and post hoc Dunn’s Test P \ 0.05). n.a. not
available
Hydrobiologia (2013) 701:129–148 141
123
upstream reaches without salinisation. This species
benefits from its physiological capacity to live in
highly saline waters (Herbst et al., 2008). Comparison
of metrics developed for classification of ecological
status under the EU-WFD, as well as other biological
metrics between sites affected by anthropogenic
salinisation and sites from the Werra and other
German rivers without intensive anthropogenic salin-
isation, revealed a severe degradation of salinisation
sites on the Werra. These sites were placed in the worst
ecological status class ‘bad’ according to the EU-
WFD. Multivariate statistical analyses revealed that
anthropogenic salinisation is still the key factor for
differences in the composition of macroinvertebrate
assemblages between reaches of the Werra with and
without salinisation. Analyses of the long-term pres-
ence–absence data of macroinvertebrates showed that
there have been no substantial improvements in the
ecological status over the past 20 years. Freshwater
tributaries did not have a positive long-term effect on
the composition of macroinvertebrate communities in
the main stream although, as a result of drift, limited
numbers of some freshwater species were found at
sites close to the mouth of the Gelster tributary. Single
specimens of freshwater macroinvertebrate taxa were
also responsible for taxa numbers larger than three (3-
halophile neozoic invertebrate species) at SS.
Although there is a possibility of recolonisation
through drift and other dispersal mechanisms, there
was no evidence that persistent populations had
established themselves. Mortality is probably high
for most drifting freshwater invertebrates or they may
be unable to complete their reproductive life cycles
after entering the salinised mainstem river, depending
0.25.1-
-1.5
2.0
Werra Reference
Thuringian Reference
Hesse Reference
Transitional Zone
Salinisation Sites Werra
Axis 1 (85.4 %)
Axis 2 (4.3 %)
Reference Sites
Transitional Zone
(Vach a)
Salinisation
sites
Reference Sites
Transitional Zone
Salinisation Sites
Fig. 4 PCA plot
(points = sample sites)
based on biological metric
composition of invertebrate
assemblages. Sample sites
are separated in the plot by
similarities and
dissimilarities in
composition of biological
metrics among invertebrate
assemblages. Envelopes
enclose reference sites, the
transitional zone and
salinisation sites
Table 6 RDA—summary of statistical measures of biological metric data and all environmental factors investigated
Axes 1 2 3 4 Total variance
Eigenvalues 0.733 0.024 0.019 0.013 1.000
Biological metric data-environment correlations 0.960 0.627 0.787 0.633
Cumulative percentage variance:
of Biological metric data 73.3 75.7 77.5 78.9
of Biological metric data-environment relationship 90.6 93.5 95.8 97.5
Sum of all canonical eigenvalues 0.809
142 Hydrobiologia (2013) 701:129–148
123
on salt loads, discharge (dilution), season and inver-
tebrate taxon (salinity tolerance, life cycle stage).
Invertebrate drift from the stretch of the Werra
unaffected by salt brine also seems to be primarily
responsible for the ‘moderate’ ecological status
observed at the TZ (Vacha).
Our results indicate that anthropogenic salinisation
in the Werra is a serious threat to the biodiversity of
stream macroinvertebrates. The same is true for other
biota (Hu
¨
bner, 2007) including fish (Wagner, unpub-
lished data) and therefore very probably for the whole
community of the river ecosystem.
Table 7 Ranking of 70 environmental variables by their effects on composition of biological metrics of invertebrate assemblages
Variable Marginal effects Conditional effects
Lambda1 LambdaA FP
K
max
0.71 0.71 132.03 0.005
K
mean
0.70
Cl
mean
0.69
Cl
min
0.65 0.00 0.92 0.430
Na
mean
0.65
Mg
min
0.65 0.00 0.27 0.890
Mg
mean
0.65
Cl
max
0.64 0.00 0.55 0.645
Na
max
0.64
Na
min
0.63
dH
mean
0.62 0.01 0.52 0.615
Mg
max
0.61 0.00 0.33 0.870
dH
max
0.59 0.00 0.61 0.660
Cl
Std.
0.57 0.00 0.29 0.850
dH
min
0.56 0.01 1.15 0.295
Ca/Mg
min
0.50
Ca
Std.
0.48 0.00 0.97 0.425
Ca/Mg
mean
0.47
Ca/Mg
max
0.47
SO
4mean
0.44
K
min
0.42
SO
4max
0.41
Mg
Std.
0.40 0.01 1.39 0.180
LF
mean
0.38
LF
max
0.37
SO
4min
0.36
SO
4Std.
0.36
LF
min
0.34 0.00 0.57 0.650
dH
Std
. 0.32 0.02 4.72 0.005
o-PO
4
–P
min
0.26
NH
4
–N
min
0.23 0.00 1.06 0.355
Ca
max
0.21
Only variables with Lambda1 [ 0.2 are shown (32 of 70 variables). Lambda1 indicates the percentage of variability explained by a
single variable. LambdaA indicates the percentage explained by a variable after the forward selection starting from the best variable
(marginal effects). Each subsequent variable is ranked on the basis of the fit that the variables give in conjunction with the variables
already selected (conditional effects). P and F values indicate the level of significance of each variable obtained by Monte Carlo
permutations (199 random permutations). Significant variables are italicised
Hydrobiologia (2013) 701:129–148 143
123
Natural geological/geogene salinisation and histor-
ical salt production at the Werra can be assumed to
have had rather marginal and local effects. This
changed with increasing industrialisation. Potassium
salts (KCl and MgSO
4
) are still the basis of potash
fertilizers, which are in demand around the world. The
consequences of anthropogenic salinisation have been
the subject of discussion since the beginning of the
nineteenth century, when industrial potash fertilizer
production was established in Germany. One of the
first notes on these discussions (Anonymous, 1918),
concerning other Thuringian rivers (the Ilm and the
Table 8 RDA—summary of statistical measures of biological metric data and two environmental factors determined by forward
selection
Axes 1 2 3 4 Total variance
Eigenvalues 0.725 0.005 0.117 0.051 1.000
Biological metric data-environment correlations 0.954 0.444 0.000 0.000
Cumulative percentage variance:
of Biological metric data 72.5 73.0 80.3 81.8
of Biological metric data-environment relationship 99.3 100.0 0.0 0.0
Sum of all canonical eigenvalues 0.710
Monte Carlo Permutation tests of significance of RDA axis 1: F = 142.22, P = 0.005; of the overall test: F = 72.84, P = 0.005)
metarhithral
(sc. Taxa)
2.2
5.1-
-1.5
1.8
Abundance
Number of Taxa
German
Saprobic
Index
Diversity
(simpson index)
Potamon
Typie index
r/K relation
metarhithral
Rheoindex
freshwater
oligohaline
mesohaline
polyhaline & euhaline
EPT-Taxa
EPT(abu. cl.)
EPTCBO
Score General
Degradation
Score German
Fauna Index
Score
metarhithral
Score EPT
Score EPT
°dH
Std.
K
+
MAX
Biological metrics
Environmental variables
Salinisation sites
Transitional zone
Reference sites(Werra and
rivers in Thurin
g
ianand Hesse)
Fig. 5 RDA triplot,
showing groups of samples
as areas (done by enveloping
samples with lines),
biological metrics (small
arrows) and the two
explanatory environmental
variables potassium
concentration maximum
(K
max
?
) and standard
deviation of total hardness
(°dH
Std.
)aslarge arrows
144 Hydrobiologia (2013) 701:129–148
123
Saale), reveals a maximum level of 450 mg l
-1
chloride
and a maximum level of total hardness \65°dH
(*1,157 mg l
-1
CaCO
3
concentration equivalent).
Thuringian law (and \45°dH Prussian law) were
considered acceptable in order to prevent damage to
human health and anthropogenic uses of the river (e.g.
fisheries). In the Werra, thresholds of 2,500 mg l
-1
for
chloride and 90°dH (*1,602 mg l
-1
CaCO
3
concen-
tration equivalent) for total hardness are currently still
in force. The current limit for chloride was adopted
from a historical limit established in 1942 as a ‘war
limit value’ to ensure fertilizer supply during World
War II. The threshold is unique in Germany and the
value is approximately 12 times higher than the
common environmental target of 200 mg l
-1
for
freshwater ecosystems in Germany (LAWA, 2007),
which was set in line with EU-WFD demands.
This common environmental target is within the
range of river-type-specific environmental targets of
other European Member States (e.g. Belgium: 150
mg l
-1
; the Netherlands: 40–400 mg l
-1
; Austria:
150 mg l
-1
; Czech Republic: 100–150 mg l
-1
;
Bulgaria: 100–150 mg l
-1
; Poland: 300 mg l
-1
). Nat-
ural Cl
-
background concentrations are estimated to be
50 mg l
-1
(LAWA, 2007) in German freshwater
ecosystems, although measurements have shown
even lower values. The first measurements at the
mouth of the River Elbe (in Hamburg) on 1 June 1852
revealed an average concentration of 23.9 mg l
-1
Cl
-,
which is approximately five times lower than today
(123 mg l
-1
) (Bergemann, 2005). Estimates of refer-
ence conditions for the lower section of the Werra with
regard to physicochemical factors were made by
Hu
¨
bner (2007) and Braukmann & Bo
¨
hme (2010).
Our results indicate that the official legal limits for
chloride concentration and total hardness described
above and current concentration levels for ions
associated with salinisation (potassium (K
?
), sodium
(Na
?
) and magnesium (Mg
2?
)) will prevent macroin-
vertebrate communities from reaching the ‘good
ecological status’ demanded by the EU-WFD.
In recent years a scientific dispute has emerged
about the relevance of factors controlling the ecolog-
ical status in salinised parts of the Werra. Braukmann
2.0-1.5
-1.5
2.0
Wernshausen
Breitungen
Barchfeld
Tiefenort
Vacha
Gerstungen
Frankenroda
Lauchröden
Frieda
Eschwege
Witzenhausen
Hann Münden
1985
1991
1995
Axis 1 (44.0 %)
Axis 2 (8.3 %)
Fig. 6 PCA plot
(points = sample sites)
based on current and the
long-term invertebrate
presence–absence data.
Sample sites (large dots) are
separated in the plot by
similarities and
dissimilarities in
assemblage composition
(small dots = taxa)
Hydrobiologia (2013) 701:129–148 145
123
&Bo
¨
hme (2010) maintain that salinisation, including
associated ions and water hardness, is the most
important factor. Ba
¨
the & Coring (2010), however,
suggest that salinisation is currently not the dominant
pressure and they highlight the role of other factors
such as organic and inorganic pollution as well as
hydromorphological degradation in shaping macroin-
vertebrate communities and consider these to be
responsible for the bad ecological status. Our inves-
tigation clearly supports the results and interpretations
of Braukmann & Bo
¨
hme (2010) and extends them.
Other anthropogenic pressures, such as hydromorpho-
logical degradation, organic pollution and high nutri-
ent levels certainly affect species composition, but the
continuing high salt concentration means that these
effects are masked by the massive salinisation effects.
Thus, anthropogenic salinisation is currently the
dominant pressure causing the strong discrepancy
between species composition in those reaches of the
Werra that are influenced by salt brine and those that
are unaffected.
Considering the current situation, with a regional
chloride limit of 2,500 mg l
-1
, and the long history of
discussions about salinisation in general, which are as
old as potash fertilizer production in this area, we do
not foresee the sustainable change in economic,
political and social interests that would be essential
before the former ‘freshwater ecosystem’ could
regenerate. Regeneration of the ecosystem will prob-
ably begin once the mineral deposits in this area have
been exhausted. The speed of this regeneration is
largely unknown, so that the Werra will probably bear
the inglorious label of ‘most disturbed European
river’, as it was recently entitled by regional newspa-
pers, for the next 100 years, and possibly much longer.
Salinisation of the Werra is a prominent example of
the conflicts between ecological costs and economic
benefits.
Acknowledgments We are grateful to the German Nature and
Biodiversity Conservation Union (NABU) and the ‘Bu
¨
ro am
Fluss—Lebendige Weser e.V.’ for financial support during this
study. The regional authorities of Hesse and Thuringia provided
data for this investigation, which significantly increases the
generalisability of the results. Special thanks go to Dr. Gerd
Hu
¨
bner, Stephan Gunkel and Dr. Claus-Ju
¨
rgen Schulz for
discussions during the planning of this study and on
interpretations of the results. Furthermore, we thank Prof. Dr.
Ulrich Braukmann for many valuable comments on an early
draft of this document. Andrew J. Davis helped improve the
English language of the manuscript. We further thank one
anonymous reviewer for many helpful comments.
References
Anonymous, 1918. Gutachten des Reichs-Gesundheitsrates u
¨
ber
den Einfl der Ableitung von Abwa
¨
ssern aus der Chl-
orkalium- und Sulfatfabrik der Gewerkschaft Rastenberg in
Rastenberg. Thu
¨
ringen auf die Ilm, Lossa und Saale. Be-
richterstatter: Geheimer Medizinalrat Prof. Dr. Fra
¨
nken,
Halle. Mitberichterstatter: Geheimer Ober Baurat Dr. Ing.
Keller, Berlin und Regierungsrat Prof. Dr. Spitta, Berlin. In:
Arbeiten aus dem kaiserlichen Gesundheitsamte. Beihefte
zu den Vero
¨
ffentlichungen des Kaiserlichen Gesundheit-
samtes: (44): 531–607. Julius Springer Verlag, Berlin.
AQEM consortium, 2002. Manual for the application of the
AQEM system. A comprehensive method to assess Euro-
pean streams using benthic macroinvertebrates, developed
for the purpose of the Water Framework Directive. Version
1.0, February 2002.
ARGE Weser, 2000. Folgen der Reduktion der Salzbelastung in
Werra und Weser fu
¨
r das Fließgewa
¨
sser als O
¨
kosystem.
ARGE Weser, Hannover.
Ba
¨
the, J. & E. Coring, 2010. Biological effects of anthropogenic
salt-loads on the aquatic fauna: a synthesis of 17 years of
biological survey on the rivers Werra and Weser. Limno-
logica 41(2): 125–133.
Ba
¨
the, J., V. Herbst, G. Hofmann, U. Matthes & R. Thiel, 1994.
Folgen der Reduktion der Salzbelastung in Werra und
Weser fu
¨
r das Fließgewa
¨
sser als O
¨
kosystem. Wasser-
wirtschaft 84: 528–536.
Bergemann, M., 2005. Berechnung des Salzgehaltes der Elbe.
Wassergu
¨
testelle Elbe: 1–8, Hamburg.
Braukmann, U. & D. Bo
¨
hme, 2010. Salt pollution of the middle
and lower section of the river Werra (Germany) and its
impact on benthic macroinvertebrates. Limnologica 41(2):
113–124.
Brock, M. A., D. L. Nielsen & K. Crossle, 2005. Changes in
biotic communities developing from freshwater wetland
sediments and experimental salinity and water regimes.
Freshwater Biology 50: 1376–1390.
Davis, J., L. Sim & J. Chambers, 2010. Multiple stressors and
regime shifts in shallow aquatic ecosystems in antipodean
landscapes. Freshwater Biology 55(1): 5–18.
European Community, 2000. Directive 2000/60/EC of the
European Parliament and of the Council of 23 October
2000 establishing a framework for Community action in
the field of water policy, No. L 327/1, of 22 December
2000.
Fesel, U., W. Klo
¨
s & W. Meinel, 1996. Vergleichende biolog-
ische Untersuchungen an der Werra (Juni und Juli 1995).
Wiesbaden: HLfU (Umweltplanung, Arbeits- und Um-
weltschutz 213).
FGG Weser, 2005. EG-Wasserrahmenrichtlinie—Bewirtschaf-
tungsplan Flussgebietseinheit Weser 2005—Best-
andsaufnahme Koordinierungsraum Werra. [Internet].
[cited 2008 September] Available from: www.fgg-weser.
de/wrrl/bericht_2005.html.
146 Hydrobiologia (2013) 701:129–148
123
Fittkau, E. J., 1950. Mitteilung u
¨
ber die in der Fulda und ihren
Zuflu
¨
ssen aufgefundenen Weichtiere. Jahresbericht der
Limnologischen Station Freudenthal 1 (1949): 17–19,
Hann-Mu
¨
nden.
Gotelli, N. J. & R. K. Colwell, 2001. Quantifying biodiversity:
procedures and pitfalls in the measurement and comparison
of species richness. Ecology Letters 4: 379–391.
Gotelli, N. J. & G. L. Entsminger. 2004. EcoSim: Null models
software for ecology. Version 7. Acquired Intelligence Inc.
& Kesey-Bear. Jericho, VT 05465. [Internet]. [cited 2009
April] Available from: http://garyentsminger.com/ecosim/
index.htm.
Haase, P., A. Sundermann & K. Schindehu
¨
tte, 2006a. Opera-
tionelle Taxaliste als Mindestanforderung an die Bestim-
mung von Makrozoobenthosproben aus Fließgewa
¨
ssern
zur Umsetzung der EU-Wasserrahmenrichtlinie in Deu-
tschland. [Internet]. [cited 2006 May]; Available from:
http://www.fliessgewaesserbewertung.de.
Haase, P., A. Sundermann & K. Schindehu
¨
tte, 2006b.
Informationstext zur Operationellen Taxaliste als Minde-
stanforderung an die Bestimmung von Makrozooben
thosproben aus Fließgewa
¨
ssern zur Umsetzung der
EU-Wasserrahmenrichtlinie in Deutschland. [Internet].
[cited 2006 May]; Available from: http://www.fliessge
waesserbewertung.de.
Herbst, D. B., M. T. Bogan & R. A. Lusardi, 2008. Low specific
conductivity limits growth and survival of the New Zea-
land Mud Snail from the upper Owens River, California.
Western North American Naturalist 68(3): 324–333.
Hu
¨
bner, D., 2007. O
¨
kologisch-faunistische Fließgewa
¨
sserbe-
wertung am Beispiel der salzbelasteten unteren Werra und
ausgewa
¨
hlter Zuflu
¨
sse. PhD Thesis, Universita
¨
t Kassel,
Witzenhausen.
Hulsch, J. & G. M. Veh, 1978. Zur Salzbelastung von Werra und
Weser. Neues Archiv fu
¨
r Niedersachsen 27(4): 367–377,
Go
¨
ttingen.
Hurlbert, S. H., 1971. The nonconcept of species diversity: a
critique and alternative parameters. Ecology 52: 577–585.
Illies, J. & L. Botosaneanu, 1963. Proble
`
mes et me
´
thodes de la
classification et de la zonation e
´
cologique des eaux cou-
rantes, conside
´
re
´
es surtout du point de vue faunistique.
Verhandlungen Internationale Vereinigung fu
¨
r theoreti-
sche und angewandte Limnologie 12: 1–59.
Jongman, R. H. G., C. J. F. ter Braak & O. F. R. van Tongeren,
1995. Data Analysis in Community and Landscape Ecol-
ogy. University Press, Cambridge.
Kefford, B. J., P. J. Papas, L. Metzeling & D. Nugegoda, 2004.
Do laboratory salinity tolerances of freshwater animals
correspond with their field salinity? Environmental Pollu-
tion 129: 355–362.
Koop, J. H. E., 1991. O
¨
kophysiologische Untersuchungen zum
Einfluß anorganischer Salzbelastung in Fließgewa
¨
ssern auf
den Stoffwechsel von Gammariden. Erweiterte Zusam-
menfassung DGL-Jahrestagung 1990: 387–391.
Koop, J. H. E., 1994. Akute und subakute Einflu
¨
sse von Kali-
umchloridbelastungen in Fließgewa
¨
ssern auf den Energi-
estoffwechsel von Gammarus pulex (L.) und Gammarus
tigrinus (Sexton). PhD Thesis, Heinrich-Heine Universita
¨
t,
Du
¨
sseldorf.
LAWA, 2002. Gewa
¨
sserstrukturkartierung in der Bundesre-
publik Deutschland—U
¨
bersichtsverfahren. Bayerisches
Landesamt fu
¨
r Wasserwirtschaft. Kulturbund—Verlag
GmbH, Berlin.
LAWA, 2007. RAKON, Rahmenkonzeption Monitoring—Teil
B Bewertungsgrundlagen und Methodenbeschreibung,
Arbeitspapier II—Hintergrund und Orientierungswerte fu
¨
r
physikalisch-chemische Komponenten.
McCabe, D. J. & N. J. Gotelli, 2000. Effects of disturbance
frequency, intensity, and areaon assemblages of stream
macroinvertebrates. Oecologia 124: 270–279.
Meier, C., P. Haase, P. Rolauffs, K. Schindehu
¨
tte, F. Scho
¨
ll, A.
Sundermann & D. Hering, 2006. Methodisches Handbuch
Fließgewa
¨
sserbewertung Handbuch zur Untersuchung
und Bewertung von Fließgewa
¨
ssern auf der EG-Wasser-
rahmenrichtlinie. [Internet]. [cited 2006 May]; Available
from: http://www.fliessgewaesserbewertung.de.
Meinel, W. & M. Barlas, 1987. Biozoenotische und hydro-
chemische Bestandsaufnahme der Werra und Ulster.
Wiesbaden: HLfU (Umweltplanung, Arbeits- und Um-
weltschutz 58).
Piscart, C., J. C. Moreteau & J. N. Beisel, 2005. Biodiversity and
structure of macroinvertebrate communities along a small
permanent salinity gradient (Meurthe River, France).
Hydrobiologia 551: 227–236.
Piscart, C., P. Usseglio-Polatera, J. C. Moreteau & J. N. Beisel,
2006. The role of salinity in the selection of biological traits
of freshwater invertebrates. Archiv fu
¨
r Hydrobiologie
166(2): 185–198.
Pottgiesser, T. & M. Sommerha
¨
user, 2004. Fließgewa
¨
sserty-
pologie Deutschlands: Die Gewa
¨
ssertypen und ihre
Steckbriefe als Beitrag zur Umsetzung der EU-Wasser-
rahmenrichtlinie. In: Steinberg, C., W. Calmano, R.
D. Wilken & H. Klapper (Hrsg.): Handbuch der Limnol-
ogie. 19. Erg.Lfg. 7/04. VIII-2.1, 1–16 ? Anhang.
Schmitz, W., 1960. Die Einbu
¨
rgerung von Gammarus tigrinus
Sexton auf dem europa
¨
ischen Kontinent. Archiv fu
¨
r Hy-
drobiologie 57(1/2): 223–225.
Scho
¨
nborn, W., 2003. Lehrbuch der Limnologie. E. Schweizer-
bart’sche Verlagsbuchhandlung (Na
¨
gele u. Obermiller),
Stuttgart.
Schulz, C. J., 2000. Auswirkungen ru
¨
ckla
¨
ufiger Salzgehalte auf
die O
¨
kologie mehrerer Fließgewa
¨
sser Nordthu
¨
ringens.
Wasser und Boden 52: 27–33.
Short, T. M., J. A. Black & W. J. Birge, 1991. Ecology of a
saline stream: community responses to spatial gradients of
environmental conditions. Hydrobiologia 22: 167–178.
Sim, L. L., J. A. Davis, J. M. Chambers & K. Strehlow, 2006.
What evidence exists for alternative ecological regimes
in salinising wetlands? Freshwater Biology 51(7):
1229–1248.
Simberloff, D., 1972. Properties of the rarefaction diversity
measurement. American Naturalist 106: 414–418.
ter Braak, C. J. F. & P. Smilauer, 1998. Canoco Reference
Manual and User’s Guide to Canoco for Windows.
Microcomputer Power, Ithaca.
Velasco, J., A. Millan, J. Hernandez, C. Gutierrez, P. Abellan,
D. Sanchez & R. Ruiz, 2006. Response of biotic commu-
nities to salinity changes in a Mediterranean hypersaline
stream. Saline Systems 2: 12.
Vogel, J. H., 1913. Die Abwa
¨
sser aus der Kaliindustrie, ihre
Beseitigung sowie ihre Einwirkung in und an den Was-
serla
¨
ufen. Verlag Borntra
¨
ger, Berlin.
Hydrobiologia (2013) 701:129–148 147
123
Williams, W. D., 2001. Anthropogenic salinisation of inland
waters. Hydrobiologia 466: 329–337.
Ziemann, H., 1971. Die Wirkung des Salzgehaltes auf die Di-
atomeenflora als Grundlage fu
¨
r eine biologische Analyse
und Klassifikation der Binnengewa
¨
sser. Limnologica 8:
505–525.
Ziemann, H., 1997. The influence of different ion ratios on the
biological effect of salinity in running waters of Thuringia
(Germany). Limnologica 27(1): 19–28.
Ziemann, H. & C. J. Schulz, 2010. Methods for biological assess-
ment of salt-loaded running waters fundamentals, current
positions and perspectives. Limnologica 41(2): 90–95.
148 Hydrobiologia (2013) 701:129–148
123
... The deposition of fine-sediments and the introduction of nutrients from agricultural run-off are considered major threats to the most endangered freshwater macroinvertebrates such as mussels (Geist and Auerswald 2007;Lopes-Lima et al. 2017). Anthropogenic (secondary) salinisation is responsible for the degradation of water quality on a worldwide scale and results in biological changes in ecosystems, mainly in freshwater biota (Bäthe and Coring 2011;Kang and King 2012;Arle and Wagner 2013;Bąk et al. 2020;Sowa et al. 2020). In addition, dioxins, pharmaceutical compounds and other chemicals can have chronic and acute effects on aquatic macroinvertebrates (Lopes-Lima et al. 2017). ...
... This study proved that the physical, chemical and hydromorphological anthropogenic transformations are important in explaining the distribution of macroinvertebrates in human-impacted Central European rivers. Our results, which showed a decrease in macroinvertebrate biodiversity in the most salinised rivers reflected by the lowest median values of the Shannon-Wiener index H' for the macroinvertebrate communities, are consistent with numerous studies that have been carried out in secondary saline rivers worldwide (Battaglia et al. 2005;Piscart et al. 2005;Bäthe and Coring 2011;Arle and Wagner 2013;Cañedo-Argüelles et al. 2013Ladrera et al. 2017). According to Pinder et al. (2005), the threshold value of salinity is 4100 mg dm −3 TDS, above which the biodiversity begins to decrease. ...
Article
Full-text available
Human activity triggers negative alternations in river habitats, including changes to the physical and chemical parameters of the water, its hydromorphological features and the introduction and spread of invasive alien species. These modifications are expected to be intensified by climate change. Eight rivers in one of the most urbanised and industrialised regions in Europe, i.e. the Upper Silesian Coal Basin, were surveyed in order to explain the impact of anthropopressure on the distribution of macroinvertebrates. Conductivity, altitude, hydromorphological transformations, hardness, the organic matter content and certain fractions of benthic sediments significantly affect (p < 0.01) the occurrence of macroinvertebrates in Central European rivers. Our results proved that the hydromorphological transformation of watercourses, which is expressed by the relevant indices, is one of the most predictive factors that contribute to the distribution of macroinvertebrates. Anthropogenic inland waters that have been salinised by the discharge of hard coal mine waters create new habitats for brackish and marine species that replace native freshwater species. An increase in salinity causes an impoverishment of macroinvertebrate biodiversity therefore all possible actions should be taken to reduce the anthropogenic salinity of inland waters. Secondary saline rivers may prove to be prescient for climate-induced changes to river macroinvertebrates.
... The permitted thresholds for ions only exist for Cl − , Mg 2+ and K + (2500 mg/L, 340 mg/L and 200 mg/L, respectively) and a total hardness of 90°dH (Regierungspräsidium Kassel, 2012). However, these values are not based on scientific data and greatly exceed natural ion concentrations in rivers (Huebner, 2007;Arle and Wagner, 2012). ...
... Serious ecological implications caused by effluents of the potash industry include the impoverishment of the biodiversity and abundance of local species and associated structural shifts in biotic communities (Huebner, 2007;Arle and Wagner, 2012;LAVES, 2019;Schulz and Cañedo-Argüelles, 2019). Increased disease rates of many fish species were identified as well as the absence or reduced number of teleosts typical for river habitats, such as rheophilic species but also generalists (Matthes and Werner, 2005;Schwevers et al., 2005;Matthes and Werner, 2012;LAVES, 2017LAVES, , 2019. ...
Article
Secondary salinization of freshwater ecosystems is of increasing globalconcern. One of the main causes are the effluents of the potash mining industry containing high concentrations of major ions (Cl⁻, Na⁺, Mg²⁺, K⁺). In Germany, the ongoing discharge of effluents into the River Werra led to a strong impoverishment of the biodiversity and abundance of local species. Young cohorts of many freshwater fish are completely absent suggesting reproductive failure under these conditions. Therefore, the aim of the study was to experimentally investigate the effects of high concentrations and imbalances of ions that are prevalent in potash mining effluents on reproductive traits of native freshwater teleosts. Sperm motility parameters of the common roach, Rutilus rutilus, and European perch, Perca fluviatilis, were assessed as well as fertilization rate, egg size, hatching, malformations and mortality of embryonic and larval stages of roach. Concentrations of the permitted thresholds (HT) and future thresholds (LT) as well as three ion solutions containing high Mg²⁺ (Mg), high K⁺ (K) and both in combination (Mg + K) were tested. Curvilinear velocity and linearity of perch spermatozoa were elevated with potentially adverse effects on fertilization success. Sperm motility parameters and fertilization rate of roach were not affected. However, egg sizes of roach were increased in all groups due to the osmotic action of ions and in LT, premature hatch was observed. Furthermore, all groups comprised a higher number of malformations including pericardial edema and spine curvatures and group HT exhibited a higher mortality rate compared to control. The results clearly demonstrated that particularly the sum of high concentrations of ions, as prevalent in HT and LT, rather than individual ion species exerts detrimental effects on early development of roach potentially increasing overall mortality under natural conditions. These results emphasize that currently permitted and future thresholds are exceeding tolerated ion concentrations.
... Solid wastes are backfilled or heaped, whereas liquids are injected into geological formations, such as dolomite layers (Braukmann and Böhme, 2011), water courses or oceans (Rauche, 2015). In Germany, the underground and superficial disposal of brines and the leaching of sodium chloride from piles have contaminated the Werra and Weser Rivers with sodium, chloride, potassium, magnesium, and calcium (Tockner et al., 2009;Braukmann and Böhme, 2011;Arle and Wagner, 2013). ...
Article
Full-text available
The covering of potash tailings piles with technosols (artificial soils) is a modern and promising method for decreasing the saline drainage of these piles. In this context, it is important to determine whether technosols have appropriate physical properties for crop growth. In evapotranspiration covers, physical properties, such as bulk density, particle size distribution, total porosity, proportion of large pores, and available water are particularly important because they allow for robust crop growth, which subsequently determines the evapotranspiration capacity. However, few studies have been performed to assess the physical properties of technosols and their ability to act as evapotranspiration covers on potash tailings piles. Therefore, the present study aims to evaluate the physical properties of four different technosols made of municipal solid waste incineration bottom ash and coal combustion residues installed on a potash tailings pile located in Heringen, Germany. The total porosity, infiltration capacity, particle size distribution, bulk density, wettability, water retention curve, pH, electrical conductivity, and water content were determined. The pH of the technosols averaged 8.5, the electrical conductivity varied from 2.8 to 3.3 mS/cm, the mean bulk density was 1.21 g/cm³, the total porosity was 52.8%, and the rate of medium pores was 13.9% of the technosol volume. On average, the coarse fraction accounted for 42% of the technosol mass, whereas the fine fraction accounted for 52% of the sand-size particles, 43% of the silt-size particles and 5% of the clay-size particles. Likewise, no wetting restrictions for the technosols were found. To conclude, the different technosols present no limitations for crop growth, although the heavy metal contents of municipal solid waste incineration bottom ash and coal combustion residues should be considered in future studies.
... Environmental risk assessments of potash mining usually describe the anthropogenic salinisation of groundwater and surface water. The problems of brine distribution in groundwater and surface water were researched by reference [1,2] in Germany, reference [3] and reference [4] in Russia. These studies discovered saline drainage filtrations of slurry storage facilities and salt tailing piles into ground water and discharges of saline groundwater into local river valleys. ...
Article
Full-text available
The environmental impact of deposit development can be indirect and can cause combined geochemical processes in ecosystems. These must be taken into consideration under environmental forecasting and environmental risk assessment. Soil degradation in the Taiga Forest is considered, within the area of Verkhnekamskoye potash deposit (Russia), as an example of such environmental transformation. Here, the mechanism and characteristics of the anthropogenic salinisation of alluvial soils under potash deposit development are newly described. It was found that there is a strong anthropogenic impact of the potash industry on valley soils where the contaminated Na-Cl groundwater discharges or is close to the surface. The valley soils are characterised by high salinity, and the sum of toxic salts in soils has reached 26%. Alluvial gley humic clay chloride saline soil (Gleyic Fluvisols (Salic, Loamic, Technic)) and secondary solonchak on alluvial humic clay soil sulfate-chloride gypsum-containing surface-gleyed (Chloridic Gleyic Fluvic Solonchak (Hypersalic, Loamic, Technic)) were formed in hydromorphic conditions. Morphological, physicochemical and mineralogical analyses were carried out. Under hydromorphic conditions, Chloridic Gleyic Fluvic Solonchak (Hypersalic, Loamic) was described to show a hydrotroillite layer and reddish-yellow iron-rich precipitates on its surface. The top soil horizon has the highest content of iron minerals (up to 84.9%) and Fe-bearing plant residues (up to 20%). Additionally, the spongy and gel-like organic materials, as well as the siliceous remains of diatoms, are enriched in Ca, Fe, Cl, K, Na, S and P. The lower soil horizon consists of black gel-like phases and hydrogen sulphide settings with a high content of plant residues. The insoluble part of the samples contains up to 84% hydrogoethite. The sources of iron in soils and bottom sediments include the iron-enriched Sheshma sediments speckled rocks, slurry material, halite wastes and soil minerals of alluvial gley soils.
... Salinization of the environment in the areas affected by the potash production industries is serious problem (Environmental Aspects of Phosphate and Potash Mining 2001). Salt tailing piles, slurry storage, and brine storage facilities are the main sources of contamination for the subsurface hydrosphere in the Verkhnekamskoe deposit (Bachurin and Baboshko 2008;Bel'tyukov 1996;Liu and Lekhov 2013;Lyubimova et al. 2016), Starobinskoe deposit (Belorus) (Kolpashnikov et al. 1979(Kolpashnikov et al. ,2010, Alsace deposit (France) (Baure et al. 2005;Lucas et al. 2010), and in the potash deposits in Germany (Luo et al. 2012;Arle and Wagner 2013). ...
Article
Full-text available
Importance to reduce permeability of rocks arises safety of mining operations and protecting the soil and underground water in the area of industrial waste storage with high contents of water-soluble salts. Disadvantage of the existing methods of grouting rocks by soluble salts is either their dissolution with changing of the composition or diluting of the contacting solution. The paper considers the composition of the grouting mixture for the deposition of insoluble calcium salts and its effectiveness in laboratory conditions. The eutonic solution of a four-component KCl–NaCl–CaCl2–H2O system was used as the initial solution, while saturated solutions of sodium sulfate and carbonate, and sulfuric acid were applied as the precipitators. Efficiency of rock grouting was estimated according to the water or solution filtration rate through the rock (sand, sylvinite, and halite) before and after calcium salt crystallization. Laboratory experiments showed that the precipitation of calcium sulfate responds to the challenge of plugging rocks. The paper proposes some recommendations of protective screens at the bed of the salt tailing piles and slurry storage facilities.
... Lagoon water system in areas such as Peene, Swina, and Dziwna (Zettler, 2008). On the contrary, in the German inland waters, this species is established in the Weser and Werra Rivers (Arle & Wagner, 2013;Bäthe, 2009;Herbst & Bäthe, 1993;Szöcs et al., 2014) and it has also been recorded in the Mulda River (GEO-Tag der Artenvielfalthttps, 2020 (Leuven et al., 2009;Nunes et al., 2015) and many amphipod species (Grabowski et al., 2007;Rewicz et al., 2015). It cannot be ruled out that this route may also play a role in the dispersion of native species. ...
Article
Apocorophium lacustre (Vanhoffen, 1911), which is a native brackish amphipod species of the North Atlantic and Baltic coasts, was recorded in the upper Oder River for the first time in 2017. Prior to that, only alien amphipod species had been found in this area. The aims of the study were to describe the distribution pattern of A. lacustre in the upper Oder River catchment, to investigate the composition and structure of the amphipod assemblages against the background of the habitat conditions and to provide a genetic identification of the species using DNA barcodes. In total, 16 sites were studied. Apocorophium lacustre was recorded in 14 of them. It was not found at one site in the Oder River and at one location in the lower Klodnica River. Apocorophium lacustre outnumbered the other amphipods in the Oder, in the mouth section of the Klodnica and at one site in the Gliwice Canal. The alien species Gammarus tigrinus Sexton, 1939 was dominant in the amphipod communities at the southernmost site in the Oder River, in the Klodnica Canal and at most sites along the Gliwice Canal. In the Klodnica River, another alien species, Dikerogammarus villosus (Sovinsky, 1894), was dominant in the amphipod fauna at two sites. While the density of A. lacustre was high in the Oder River, it was much lower in both the canals and the Klodnica River. In our study, the depth and river velocity both contributed to the distribution of this species. This article is protected by copyright. All rights reserved.
... These patterns were still existent in the remaining period of 2001, but in a more narrow range of oxygen concentrations. In addition, the concentrations of oxygen were validated for the mouth with data from [35]. ...
Article
Full-text available
Hasel is considered a moderately polluted river in Germany. This study investigated its water quality, examining Dissolved Oxygen (DO) and dissolved substrate (COD) with the use of AQUASIM. The calibration procedure used observed data from various locations along the river. The model’s calibration was used to study the response of Hasel River to the effluents of wastewater treatment plants and sewer overflow emissions. Results revealed that high emissions from sewerage systems may reduce the oxygen concentration to low levels. Furthermore, joined sewer overflows may disrupt the oxygen levels for a long period. In addition, oxygen was over saturation in some periods of the calibration period. The proposed model can be utilized in future analyses, improving the functional understanding of ecological processes in rivers and the identification of ecological effective management strategies.
... We found that communities at sites with higher EC change were dominated by the amphipods Gammarus tigrinus, Chelicorophium curvispinum and the snail Potamopyrgus antipodarum. These results are consistent with previous studies (Arle and Wagner, 2013;Szöcs et al., 2014) for the River Werra, Germany. Piscart et al. (2011Piscart et al. ( , 2005 also observed a similar pattern along a salinity gradient in the Meurthe River, northeastern France. ...
Article
Rising salinity in freshwater ecosystems can affect community composition. Previous studies mainly focused on changes in freshwater communities along gradients of absolute levels of electrical conductivity (EC). However, both geogenic and anthropogenic drivers contribute to the EC level and taxa may regionally be adapted to geogenic EC levels. Therefore, we examined the turnover in freshwater invertebrates along gradients of anthropogenic EC change in two regions of Germany. The anthropogenic change of EC was estimated as the difference between the measured EC and the modeled background EC driven by geochemical and climate variables. Turnover in freshwater invertebrates (β-diversity) was estimated using the Jaccard index (JI). We found that invertebrate turnover between EC gradient categories is generally greater than 47%, with a maximum of approximately 70% in sites with a more than 0.4 mS cm−1 change compared to the baseline (i.e. no difference between predicted and measured EC). The invertebrates Amphinemura sp., Anomalopterygella chauviniana and Leuctra sp. were reliable indicators of low EC change, whereas Potamopyrgus antipodarum indicated sites with the highest EC change. Variability within categories of EC change was slightly lower than within categories of absolute EC. Elevated nutrient concentrations that are often linked to land use may have contributed to the observed change of the invertebrate richness and can exacerbate effects of EC on communities in water. Overall, our study suggests that the change in EC, quantified as the difference between measured EC and modeled background EC, can be used to examine the response of invertebrate communities to increasing anthropogenic salinity concentrations in rivers. However, due to the strong correlation between EC change and observed EC in our study regions, the response to these two variables was very similar. Further studies in areas where EC change and observed EC are less correlated are required. In addition, such studies should consider the change in specific ions.
Article
Full-text available
The New Zealand mud snail (Potamopyrgus antipodarum; NZMS) is among the most globally widespread aquatic invaders, occurring in 39 countries and 5 continents. Herein we provide a systematic review of 245 articles, focusing on the ecological impacts, spatial distribution, population dynamics, vectors of spread, and management of invasive NZMS. Most NZMS introductions originate from already-established invasive populations, which represent a small number of clonal lineages. The invasion success of NZMS stems from opportunistic traits, and while their tolerance of broad ranges of environmental conditions facilitates spread, optimal conditions for successful NZMS establishment are evident: stable hydrology, slow water velocity, high specific conductivity, and moderate salinity. NZMS can become exceptionally abundant, driving the greatest secondary-production rates reported for any stream invertebrate. However, NZMS populations fluctuate seasonally and over longer time scales, with marked declines observed after population booms. Minimal genetic variation within and among invasive populations and minimal incidences of predation/parasitism suggest that environmental factors constrain populations. As detritivore-herbivores, NZMS impact multiple compartments of aquatic ecosystems and their functioning. NZMS alter invertebrate and algal communities and can resist digestion by many fish species, reducing fish condition. This lack of digestion combined with expanding NZMS populations suggest that snail-eating fish are unlikely to regulate NZMS populations and may aid in local range expansion. Management programs and technologies have recently emerged to assist resource managers, including advances in environmental DNA detection methods and effective chemical decontamination treatments. The objective of this review is to contribute to a more robust understanding of the global NZMS invasion, such that undesired impacts can be minimized or averted.
Article
Full-text available
Potash mining was carried out in Northern Thuringia for almost a century, leading to the salinization of several running waters. After mining was terminated in the 1990s, overall salt loads decreased markedly, but did not disappear completely. The consequences of salinization for the various aquatic organisms differ. It seems likely that bacterial carbon degradation remains unaffected. This might also hold true for the phytoplankton of the running water of the river Wipper. With regard to the benthic diatom communities, we found that changes in the diatom community structure of the creek Urbach after the sudden termination of salt brine inputs were restricted to half a year. It also became apparent that the strong salt load had been masking the influence of wastewater inflows. Despite increasing salinities, diatom species numbers along the saline river Wipper were remarkably constant, although changes in the species composition were visible. With respect to aquatic insects, the recolonization of formerly desolated parts of this river has given rise to a community whose members tolerate a comparatively broad spectrum of salinity.
Chapter
Die Gewässertypologie hat als angewandte limnologische Fachrichtung die Aufgabe, die ungezählten individuellen Wasserläufe nach Gemeinsamkeiten zu überschaubaren Einheiten, den Typen, zusammenzufassen. In diesem Beitrag werden die Beschreibungen der Typen, die für die Bundesrepublik Deutschland abgeleitet wurden, erstmals in gedruckter Fassung in Form von Steckbriefen vorgestellt. Sie stellen die “offizielle” deutsche Fließgewässertypologie als wichtiges Instrument für die Umsetzung der EU-Wasserrahmenrichtlinie (WRRL) und eine Vielzahl anderer Anwendungszwecke dar.
Article
The effect of salinity on aquatic organisms can be interpreted as a result of the combined action of different ions or ion ratios and the total salt concentration and may be defined as the biological effect of salinity. This biological effect of the salinity can be described by the halobion index, which may be determined on the basis of the diatom assemblages. The halobion indices from running waters of Thuringia with different salt contents were determined and the correlation between the primary effective factors of the salt content and the halobion indices were calculated for individual types of waters. Defined ion combination are efficient in the different concentration degrees of the waters. So the pH value as a integrating factor of the Ca(HCO3)2/CO2 buffering system is efficient in acidic waters with a poor content of electrolytes. In oligohalobic waters up to the limit of saline waters the combination of salinity and alkali-calcium-ratio or the "P value" is significant for the biological effect. The correlation between chloride concentration and P value and its importance to the biologial effect of salinity is demonstrated for the river Weser. These relationships can be applied to describe the conditions for the limit between the limnetic and the saline degree. Saline waters can be characterized as alkali chloride waters, in which the biological effect of salinity (osmotic effect) will be varied by specific ion actions, for instance of the potassium. This is demonstrated by a comparison between the rivers Werra and Wipper.