Content uploaded by Stefan Lorenz
Author content
All content in this area was uploaded by Stefan Lorenz on Apr 09, 2019
Content may be subject to copyright.
RESEARCH PAPER
Meander reconnection method determines restoration
success for macroinvertebrate communities in a
German lowland river
Stefan Lorenz
1,2
, Marc Leszinski
1
and Daniel Graeber
3
1
Leibniz Institute of Freshwater Ecology and Inland Fisheries Berlin, M€
uggelseedamm, Berlin
2
Julius K€
uhn Institute, Institute for Ecological Chemistry, Plant Analysis, and Stored Product Protection,
K€
onigin-Luise-Straße, Berlin
3
Department of Bioscience, Aarhus University, Vejlsøvej, Silkeborg
Re-meandering of degraded rivers is a frequently implemented measure in river restoration.
A simple solution is reconnection of old meanders; however, its success likely depends on
the reconnection method. We conducted a field study to analyze the benefits of a fully
reconnected (fully opened meander, blocked main channel) and a partially reconnected
meander (opened downstream, pipe bypass from main channel upstream, still open main
channel) for macroinvertebrate communities in a German lowland river. Immediately upon
reconnection of the two meanders, habitat diversity, and macroinvertebrates were recorded
for three years with sampling in spring and in summer each year. The results of a principal
response curve analysis show that the macroinvertebrate community in the fully reconnected
meander reflected main channel reference conditions after 1.5 years. The macroinvertebrate
community composition was not improved relative to in-stream reference conditions within
the partially reconnected meander, which could be attributed to the almost complete lack of
flow changes that resulted in missing improvements of substrate diversity. Our results show
that the meander reconnection method must sufficiently affect the basic hydromorphological
requirements to achieve reference macroinvertebrate community composition. Measures
including hydromorphological conditions are therefore recommended for employment in
environmental management.
Received: October 14, 2015
Revised: March 18, 2016
Accepted: April 27, 2016
Keywords:
Flow regime / Hydromorphology / Re-meandering / River rehabilitation / Substrate
heterogeneity
1 Introduction
Several centuries of human activities have led to
considerable changes in river hydromorphology and,
consequently, the natural character of rivers. Nowadays,
rivers worldwide are hydromorphologically degraded due
to, for instance, water diversions, channelization, or
habitat and flow alterations (USEPA, 2009; EEA, 2012;
Feh
er et al., 2012). These hydromorphological changes
have far-reaching consequences for biodiversity and river
ecosystem functioning (e.g., Elosegi and Sabater, 2013).
Hence, river restoration aims to rebuild the former pristine
river morphology of most lowland rivers.
Multiple measures are applied in river restoration,
including re-meandering, removal of bank fixation, and
habitat improvement (Brookes and Shields, 1996). Re-
meandering, for example by reconnection of former
meanders or oxbow lakes, is thus only one on various
measures, but it promises the greatest success as it may
Handling Editor: Ralf Ibisch
Correspondence: Stefan Lorenz, Leibniz Institute of Freshwater
Ecology and Inland Fisheries Berlin, M€
uggelseedamm 301, 12587
Berlin
E-mail: stefan.lorenz@julius-kuehn.de
Fax: þ0049 0 3083042503
Abbreviations: CWD, coarse woody debris; POM, particulate
organic matter; PRC, principal response curve
International Review of Hydrobiology 2016, 101,1–9 DOI 10.1002/iroh.201501823
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1
affect about 70% of all the hydromorphological processes
and variables acting in rivers (Lorenz et al., 2016). The
complex hydromorphology of meanders may significantly
improve macroinvertebrate diversity at small spatial scales
at the river-bed (Garcia et al., 2012), and studies have
shown the fish fauna to be more diverse following
hydromorphological restoration (e.g., Cederholm et al.,
1997; Roni et al., 2006; Shields et al., 2007). Thus,
meander reconnection could be a useful approach for
water managers to obtain higher biodiversity and species
richness of fish and macroinvertebrate communities in
rivers.
All restoration measures can be implemented in a
variety of ways, but only the option reducing most
efficiently the stressor effects on the biota is of interest
to environmental managers. Many restoration measures
improve river habitat structures but do not reduce the
effects of central stressors on the biota, such as altered
flow regime. Thus, often no effects of restoration are found
on both fish and macroinvertebrates (Hamilton, 1989;
Lepori et al., 2005; Roni et al., 2006). In recent years, it has
been realized that richness and diversity of aquatic fauna
and flora only improve if the natural flow regime is restored
(e.g., Roni et al., 2008; J€
ahnig et al., 2010; Palmer et al.,
2010; Bernhardt and Palmer, 2011). Meanders increase
habitat heterogeneity at the same time as they provide a
large range of hydraulic conditions (Garcia et al., 2012).
Habitat heterogeneity and biodiversity are closely linked
(e.g., Buss et al., 2004) which is also known for hydraulic
diversity and species richness (Cardinale et al., 2005).
Thus, an increase in biodiversity and species richness
might be expected when former meanders are recon-
nected to the river channels. Particularly rheophilic
species (of which many are Ephemeroptera, Plecoptera,
and Trichoptera (EPT) taxa) should benefit from recon-
necting former meander as their hydraulic requirements
could be better met in environments with variable flows
(e.g., Usseglio-Polatera, 1994; Cardinale and Palmer,
2002; Paillex et al., 2009). Hence, for the purpose of this
paper, restoration success is defined as recolonization of
rheophilic species in the reconnected sections resulting in
a higher species richness. We assume that macro-
invertebrate composition shifts to a community typical
for lowland rivers by this recolonization process, which is
indicated by a community composition similar to in-stream
reference conditions.
In this context, we conducted a field study to analyze
the benefits of two different meander reconnection options
for macroinvertebrates in the same lowland river: (I) full
reconnection by blocking the main channel with a gravel
bank and fully opening the formerly disconnected mean-
der, and (II) partial reconnection with an opening of the
meander only downstream, a bypass with a pipe
upstream, and a still open main channel. Both options
aimed at improving hydromorphology and habitat compo-
sition to allow recolonization of rheophilic macroinverte-
brates in the meanders. We hypothesized that only a
fully reconnected meander would allow recolonization of
rheophilic macroinvertebrates since partial reconnection
does not fully restore the typical flow regime and thus the
habitat composition of a lowland river.
2 Materials and methods
2.1 Study area
The River Spree is a 380 km long lowland river in northern
Germany originating in the Lusatian mountains (Saxony,
Germany). Various hydro-engineering measures led to an
extreme degradation of the River Spree until the 1960s. In
order to facilitate flood protection, generation of energy by
water mills, agricultural use of the wetlands, and shipping,
the river was channelized by shore fortifications and
meander cuttings. The mean discharge was artificially
increased by the removal of groundwater during upstream
open-cast lignite mining activities (Pusch and Hoffmann,
2000). After shutting down most of the lignite mining,
river water is now used to fill the mining holes. As a
consequence, the river has lost most of its natural
variability of the flow regime, sediments, and morphology,
leading to subsequent degradation of the riverine biota
(Pusch and Hoffmann, 2000).
The study was conducted in the M€
uggelspree, a
eutrophic 6th-order river section of the River Spree. This
section is situated in the south-east of Berlin and has been
exposed to the same hydro-engineering measures as
the other parts of the Spree. The straightening of the
River Spree resulted in a trapezoid channel profile of
the M€
uggelspree with a mean slope of 0.015%, a mean
water depth of 1.25 m at medium discharge, and a mean
channel width of 25 m (Schulz et al., 2003). Between 1998
and 2002, discharge ranged between 2.5 m
3
s
1
in summer
and 30 m
3
s
1
in early spring (Schulzet al., 2003). Presently,
the M€
uggelspree is dominated by macrophytes (K€
ohler and
Hoeg, 2000) and the midstream riverbed is covered by
shifting sand, whereas the lateral parts show also stable
sand (Schulz et al., 2003). Total nitrogen concentrations
range between 0.7 and 3.4mg N L
1
and total phosphorus
concentrations between 70 and 180 mgPL
1
, respectively
(Schulz et al., 2003).
2.2 Meander reconnection
In 2004 and 2005, an intensive restoration program was
implemented in the M€
uggelspree. The main objective was
to reconnect two previously cut meanders to the main
channel in order to restore the typical hydromorphology of
S. Lorenz et al.International Review of Hydrobiology 2016, 101,1–9
2 © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
a lowland river via two initiatives: I) The first meander (MI)
was reconnected in 2005 by opening completely the
former course of the river, while simultaneously blocking
the former main channel by a gravel dam to redirect all flow
through the new meander (Fig. 1). The remaining old main
stem stretch served as a new flow-protected habitat (new
oxbow). II) The second meander (MII) was reconnected in
2004 by installing a bypass with a tube in the dam that
historically cut the meander. The main channel was not
blocked, allowing flow in both the main channel and the
reconnected meander (Fig. 1). Due to the combination of
the bypass and the open main channel for MII, the main
water current remained in the former main channel and not
in MII, where only a marginal proportion of discharge was
diverted to. At both sites, the old meanders were dredged
prior to reconnection.
2.3 Sampling sites
Macroinvertebrates were sampled and habitat character-
istics recorded at five 100-m sections in the M€
uggelspree
(Fig. 1). At each reconnected meander, one sampling site
was located directly below the opened site (referred to as
MIa and MIIa), while a second sampling site was located at
each meander in the mid-section of the meander course
(referred to as MIb and MIIb). An in-stream reference site
located near the villages of Wulkow and M€
onchwinkel was
sampled upstream of both meanders. This reference site
(in close proximity to the reconnected meanders) demon-
strated morphological condition that showed only minor
hydromorphological deficits such as remains of former
shore fortifications (boulders of former groyne heads) and
incomplete grove stands at the shorelines.
2.4 Habitat diversity
Habitat distribution was recorded at each sampling site in
five cross-section profiles in April and August from 2005 to
2007. The habitat types present were stable sand, shifting
sand, gravel, particulate organic matter (POM), coarse
woody debris (CWD), stones, submerged macrophytes,
and submerged tree roots. For each full cross-section
meter, the dominant habitat type was recorded. Subse-
quently, the relative share of each habitat at each cross-
section was calculated. Flow velocity was measured 10 cm
above gravel substrate at 3 points at each section using an
electromagnetic flow sensor (Nautilus C2000 Current
Meter, Ott Messtechnik, Kempten, Germany).
2.5 Macroinvertebrate sampling
Macroinvertebrates were sampled at each sampling site
in April and August from 2005 to 2007. In 2007, sampling
was only undertaken at the reference site and within
the MI meander. Stable sand and POM were sampled
using a Birge-Eckman grab sampler (six samples of
0.0225 m
2
). Shifting sand and gravel were sampled using
a Surber sampler (three samples of 0.09 m
2
, 250-mm
mesh). CWD and stones (nine samples) were sampled
using a Surber sampler (250-mm mesh) and subsequently
brought ashore, and adhering invertebrates were carefully
scraped off with a toothbrush into water-filled boxes. The
surface area of stones and CWD was calculated after
invertebrate sampling. Submerged macrophytes and roots
of riparian trees (nine samples) were sampled with a hand
net (250-mm mesh, width 24 cm), and the size of the
sampled area was estimated by multiplying hand net width
with sampling depth. The samplings of each habitat type
were evenly distributed over the 100-m sampling section
and the samples were taken at varying water depths and
at varying water flows. The samples of each habitat type
were subsequently pooled. The total area sampled was
0.135 m
2
(stable sand, POM), 0.27 m
2
(shifting sand,
gravel), 1.22 0.15 m
2
(CWD), 0.89 0.07 m
2
(stones),
0.5625 m
2
(roots), and 2.25 m
2
(macrophytes). Samples
were sieved in the field (350-mm mesh) and conserved
in 98% ethanol. Macroinvertebrates were sorted in the
laboratory using binocular microscopes and identified to
Figure 1. Map of the meanders and reference site after
reconnection to the River Spree with flow directions of the
river. Meander MI is fully connected and the former main
channel was blocked, meander MII is partially connected
as it was opened at one site, the other site was connected
via a tube (symbolized in the map) to the main channel and
the main channel itself was left open. The map is based on
two aerial photos from 2002 and 2009 (Google Earth,
Google Inc., Mountain View, USA).
International Review of Hydrobiology 2016, 101,1–9 Meander reconnection method
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 3
species level, except for Diptera (family level), Hydrach-
nidiae (subcohort level), and Oligochaeta (class level).
The complete species inventory of all sample years is
available upon request to the corresponding author.
2.6 Data analysis
We calculated average species abundances of each
sampling site by dividing the relative coverage of the
respective habitat in the five cross-sections by the relative
proportion of the sampled habitat surface, and subse-
quently multiplying this factor with the respective habitat-
specific abundance of each species. Subsequently, we
summed habitat-specific abundances of each species and
calculated the number of individuals per square meter.
For each sampling site, we calculated the biotic metrics
number of taxa, number of EPT taxa, share of rheophilic/
rheobiont taxa (Schmedtje and Colling, 1996; see SI
Table 1 for the classification of current preference groups
of all taxa), and Evenness. The share of rheophilic/
rheobiont taxa was calculated based on abundance
classes (AC): AC 1 ¼1–2 individuals m
2
,AC2¼3–
10 individuals m
2
,AC3¼11–30 individuals m
2
,AC
4¼31–100 individuals m
2
,AC5¼101–300 individuals
m
2
,AC6¼301–1000 individuals m
2
,AC7>1000 indi-
viduals m
2
. The Shannon–Wiener-Index (SWI) of habitat
heterogeneity was calculated using average relative
shares of habitat proportion based on the five measured
cross-sections (see also above, “Habitat diversity'').
A principal response curve (PRC) was used to
investigate the effect of the full or partial reconnection of
the meanders MI or MII on macroinvertebrate community
composition, respectively. PRCs are a multivariate statis-
tic used to investigate whole-community responses and to
assess which species are mostly responsible for the
change in community composition (see Van den Brink and
ter Braak (1999) for details). Here, we used a PRC based
on Bray-Curtis similarities as this better reflects differ-
ences in species composition than Euclidean distances
and has been used in previous studies assessing
macroinvertebrate communities (Hille et al., 2014). The
function capscale (Anderson and Willis, 2003) from the
vegan package (Oksanen et al., 2013) in R(version 3.1.2,
R Core Team, 2014) was used to calculate the scores for
the PRC based on Bray-Curtis similarities, but with the
same model structure as originally proposed (Van den
Brink and ter Braak, 1999). To test if the PRC model
including re-meandering type significantly explained the
changes in species composition, a permutative analysis of
variance (ANOVA) was conducted with 9999 iterations
(anova.cca function from the vegan package, Van den
Brink and ter Braak, 1999; Legendre et al., 2011).
In addition to the treatment effect, it was tested whether
time (using all data from 2005–2007) had a significant
effect on macroinvertebrate community composition for
either the fully or partially reconnected meander (sampling
sites within the same meander were pooled for analysis).
For that, a permutative multivariate analysis of variance
(PERMANOVA) was conducted with 9999 iterations using
the adonis function from the vegan package and the sites
within the meanders as blocks (Anderson, 2001).
3 Results
3.1 Habitat diversity
The sediment bed of the fully connected meander MI
consisted mainly of POM after dredging and reconnection
in 2005 (Table 1). The relative proportion of stable habitats
(stable sand, gravel, stones) increased consistently from
2005 to 2007, and SWI of habitat heterogeneity equaled
Table 1. Relative proportion (%) of habitat types sampled from 2005 to 2007
2005 2006 2007
RMIa MIb MIIa MIIb RMIa MIb MIIa MIIb RMIa MIb MIIa MIIb
St. sand 9.6 1.9 0 1.2 0 8.1 23.0 23.8 0 0 4.9 18.9 20.1 0 0
Sh. sand 26.1 0 0 0.8 0 21.0 43.1 34.3 0 0 22.1 40.3 27.8 0 0
Gravel 10.6 0 0 0 0 15.8 5.7 1.6 0 0 19.9 7.4 6.3 0 0
POM 16.4 72.7 76.9 68.7 67.0 20.5 4.8 15.3 62.0 70.1 24.1 8.6 21.4 80.8 73.3
CWD 4.6 5.7 4.8 4.7 18.1 6.7 3.2 4.4 5.3 15.7 4.1 2.2 3.1 12.3 16.7
Stones 3.3 4.1 5.4 0 1.0 7.1 6.5 10.1 0 0 4.7 10.7 11.1 0 0
Roots 2.1 1.5 1.4 1.7 3.3 2.3 2.4 2.0 1.8 3.6 1.3 1.1 1.1 2.8 2.2
Macr. 27.3 14.1 11.5 22.9 10.6 18.5 11.3 8.5 30.9 10.6 18.9 10.8 9.1 4.1 7.8
R, reference site; MIa/MIb, sampling sites in the fully reconnected meander; MIIa/MIIb, sampling sites in the partially
reconnected meander; st. sand, stable sand; sh. sand, shifting sand; POM, particulate organic matter; CWD, coarse woody
debris, macr., macrophytes.
Values of cross-section profiles have been averaged over April and August.
S. Lorenz et al.International Review of Hydrobiology 2016, 101,1–9
4 © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
reference conditions in August 2006 (Fig. 2). The relative
proportion of habitat types changed only marginally over
time in meander MII, which was connected by a bypass
with the main channel (Table 1). Similarly, SWI of habitat
heterogeneity remained constantly below reference con-
ditions (Fig. 2). The main change in meander MII was an
increase in the relative share of POM, indicating siltation of
the meander despite reconnection. This was confirmed
by flow measurements, showing that flow velocity did not
increase in meander MII (Table 2). Contrastingly, in
meander MI flow velocities were comparable to those at
the reference site in April 2006 (Table 2, not measured in
August 2005).
3.2 Macroinvertebrate community
Meander type (partial reconnection, full reconnection) had
a significant effect on macroinvertebrate community
composition relative to the reference site, as shown by
the first axis of the PRC (F¼5.9, Df ¼1, p<0.001,
permutative ANOVA of the first axis of the PRC model,
Fig. 3). The other PRC axes revealed no significant effects
and are thus not shown here. Meander type explained
43.3% of the total variance in macroinvertebrate species
composition. Another 33.3% of the total variance owed to
the effect of time. Hence, 23.5% of the variance remained
unexplained by the PRC model.
Only four species (Caenis robusta,Cloeon dipterum,
Plea minutissima, and Cyrnus crenaticornis) responded
negatively to the meander reconnection, whereas 24
species (11 Trichoptera and 3 Ephemeroptera species)
responded positively (Fig. 3). The strongest positive
response was exhibited by Baetis rhodani,Hydroptila
sp, Chelicorophium curvispinum,Pisidium sp, Athrips-
odes cinereus, and Gammarus tigrinus (Fig. 3).
The fully reconnected meander sites (MIa and MIb)
exhibited a significant effect of time (PERMANOVA,
p¼0.007, R
2
¼0.298), showing environmental improve-
ments nearing the conditions prevailing at the reference
site (Fig. 3). In contrast, the partially reconnected meander
Figure 2. Shannon–Wiener-Index (SWI) of habitat het-
erogeneity based on the average relative shares of habitat
proportion. Ref, reference; MIa & MIb, sampling sites in the
fully reconnected meander; MIIa & MIIb, sampling sites in
the partially reconnected meander.
Table 2. Flow velocity in cm s
1
(mean SD) recorded at the sampling sites
2005 2006 2007
April August April August April August
Reference 44.9 4.1 23.9 3.6 33.7 5.3 11.1 7.8 33.8 5.8 25.1 3.4
MI 0 n.d. 39.8 2.0 9.9 2.3 33.8 4.4 19.0 3.4
MII 2.8 1.8 0 1.8 1.4 0 0 0
MI, fully reconnected meander; MII, partially reconnected meander; n.d., not determined; zero values indicate no flow/
standing water.
Measurements of both sampling locations in each meander have been averaged. Flow measurements in April 2005 have
been conducted before reconnection of the meanders.
Figure 3. Principal response curve for MIa & MIb
(sampling sites in the fully reconnected meander) and
MIIa & MIIb (sampling sites in the partially reconnected
meander). The curves represent the development of the
macroinvertebrate community relative to the control
(reference site). On the right hand side, species weights
are given to show their correlation with the curves.
International Review of Hydrobiology 2016, 101,1–9 Meander reconnection method
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 5
sites did not exhibit a significant effect of time (p¼0.160,
R
2
¼0.165, MIIa and MIIb, Fig. 3).
No clear effect of meander reconnection type could be
discerned concerning taxa number relative to in-stream
reference conditions (Fig. 4a). In contrast, the number of
EPT taxa in the fully reconnected meander (sampling sites
MIa and MIb) increased more than the number of EPT taxa
in the partially reconnected meander (sampling sites MIa
and MIb) and closely resembled the number of EPT taxa in
the reference site in 2007 (Fig. 4b). Similarly, the share of
rheophilic/rheobiont taxa was similar to that at the
reference site in the fully reconnected meander (MIa
and MIb) already in 2006, while the share of rheophilic/
rheobiont taxa at the partially reconnected meander (MIIa
and MIIb) consistently decreased during the entire study
period relative to in-stream reference conditions (Fig. 4c).
4 Discussion
The aim of this study was to detect whether two types of
meander reconnection –a full or a partial reconnection of
previously cut meanders to the river –had different effects
on benthic macroinvertebrate communities. Our results
supported our hypothesis that only a fully reconnected
meander allows recolonization of rheophilic macroinverte-
brates. Thus, full reconnection resulted in a significant
increase in habitat heterogeneity and flow velocity. These
environmental improvements were clearly linked to
recolonization of a rheophilic macroinvertebrate commu-
nity with a community composition similar to that found at
the reference site. Our results also show that partial
reconnection of a cut and closed meander entails almost
no or only very little improvement of flow velocity and
habitats, thus having a negligible effect on the macro-
invertebrate community.
The macroinvertebrate community of both meanders
exhibited considerable differences in species composition
directly after reconnection with the main channel (Fig. 3).
The subsequent significantly increasing similarity between
the species composition at the fully reconnected meander
and the reference site until 2007 can be attributed to a
pronounced species-turnover. Contrarily, no significant
change in species composition over time was revealed for
the partially reconnected meander. Biotic improvement in
the same fully reconnected meander was also previously
Figure 4. Number of taxa
(a), number of EPT (Ephem-
eroptera, Plecoptera, and
Trichoptera) taxa (b), and
share (%) of rheophilic/
rheobiont taxa (c). All values
are plotted relative to in-
stream reference condi-
tions. Ref, reference, MIa
& MIb, sampling sites in the
fully reconnected meander,
MIIa & MIIb, sampling sites
in the partially reconnected
meander.
S. Lorenz et al.International Review of Hydrobiology 2016, 101,1–9
6 © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
reported for aquatic macrophytes (Gr€
unert et al., 2007).
However, full reconnection of the meander appeared
ineffective to improve the fish community as the main
habitat bottlenecks for fish are still present (Wolter, 2010).
In 2007, the species composition of the fully recon-
nected meander was already similar to that of the
reference site (Fig. 3). Therefore, the full reconnection
of the meander succeeded in restoring the macroinverte-
brate community to in-stream reference conditions already
after 1.5 years. Furthermore, rheophilic species, espe-
cially of the classes Ephemeroptera and Trichoptera, were
positively affected in the fully connected meander, which
thus is able to support a larger range of these taxa (Fig. 3,
Fig. 4b, 4c). Ephemeroptera such as Baetis rhodani and
Trichoptera such as Athripsodes cinereus and Hydroptila
sp. that showed strongest positive responses to meander
reconnection in our study were likely influenced by the
improvement in available substrate types (Timm et al.,
2008), but macroinvertebrates have generally been shown
to profit from a diverse substrate heterogeneity (Buss
et al., 2004; Garcia et al., 2012). River stretches with areas
containing larger substrates support a higher number of
macroinvertebrates than stretches consisting mainly of
uniform small substrates (Duan et al., 2009; Pan et al.,
2012). Establishment of diverse substrate heterogeneity is
therefore a prerequisite for improving macroinvertebrate
communities in connection with restoration (Buss et al.,
2004), as shown previously for fish at the same study sites
(Wolter, 2010). As flow velocity only changed negligibly
in the partially reconnected meander compared to the
fully reconnected meander, improvement of substrate
heterogeneity did not occur. The failure of the partially
reconnected meander to restore the rheophilic macro-
invertebrate community may thus be attributed to the
almost complete absence of changes in flow velocity
(Roni et al., 2008; J€
ahnig et al., 2010; Palmer et al., 2010;
Bernhardt and Palmer, 2011).
Species with high dissolved oxygen requirements,
such as Baetis sp. that responded positively to meander
reconnection, actively search for favorable microhabitats
(Lowell and Culp, 1999), which are provided by the
changes in flow in the fully reconnected meander. Vice
versa, hydromorphological degradation of the River
Spree section, which led to flow reduction and subsequent
dissolved oxygen depletion, altered the macroinvertebrate
communities in the opposite way (Graeber et al., 2013).
Furthermore, suspension feeders such as Pisidium sp.,
Hydropsyche pellucidula, and Unio tumidus in our study,
depend on flow (e.g., Growns and Davis, 1994; Chester
and Robson, 2011); thus, the positive response of these
species to meander reconnection probably resulted from
the increased food availability by flow increases. The
strong positive response of Chelicorophium curvispinum
and Gammarus tigrinus to meander reconnection is likely
attributed to the colonization of previously isolated habitats
by these highly invasive species (Grabowski et al., 2007;
Noordhuis et al., 2009).
Species that responded negatively to the reconnection
of meanders, such as Caenis robusta and Cyrnus
crenaticornis, prefer still or slow flowing water with large
amounts of debris (Bradbeer and Savage, 1980; Reusch
and Brinkmann, 1998). As flow increased in the fully
reconnected meander, such micro-habitats can only
establish in specific areas of the restored site at lowered
coverage resulting in lower abundances of species
depending on these micro-habitats. Similarly, species
that prefer to colonize still or slow flowing water and
that are capable of tolerating low dissolved oxygen
concentrations, such as Cloeon dipterum (Nagell and
Fagerstr€
om, 1978), also responded negatively to meander
reconnection in our study as the extend of such micro-
habitats likewise decreased. Additionally, species that
responded negatively to the reconnection of meanders
may have especially suffered from river dredging prior to
restoration.
Meanders provide suitable habitats both for lotic and
lentic species by creating a high habitat heterogeneity that
includes also flow refugia and hydraulic dead zones which
are essential to sustain overall species richness at the
floodplain scale (Garcia et al., 2012). Therefore, recon-
nection of meanders or creation of new meanders does
have the ability to restore the macroinvertebrate commu-
nity typical of lowland rivers (including lotic and lentic
species), but only if the re-meandering allows for a natural
hydromorphology. However, a reduction of biodiversity
resulting from losses of lentic specialist may occur at
the floodplain scale when former isolated habitats are
reconnected to the main channel (Paillex et al., 2009).
Depending on the specific restoration target, which might
be different for preserving/creating a maximum of
biodiversity or for restoring a more natural community,
such losses may be either acceptable or not.
Habitat restoration (including re-meandering) acts on
much larger time scales than covered by our study period
(Lake et al., 2007; Verdonschot et al., 2013). However, the
results of our study showed that re-meandering can also
lead to short-term ecological improvements, if performed
adequately. Here, the observed rapid colonization of
macroinvertebrates in the fully reconnected meander is
well in line with the findings of several restoration studies
(e.g., Biggs et al., 1998; Laasonen et al., 1998), but rare
species often appear at later stages of the colonization
process (e.g., Biggs et al., 1998). Accordingly, the initial
succession of macroinvertebrates is mainly related to the
biological drift of upstream fauna (Matthaei et al., 1997). At
small river stretches such as the ones studied here, habitat
restoration has previously been reported to be insufficient
for improving macroinvertebrate community composition
International Review of Hydrobiology 2016, 101,1–9 Meander reconnection method
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 7
(J€
ahnig et al., 2010). However, our results show that the
small-scale meander reconnection of a river section of
several hundred meters may, in fact, improve macro-
invertebrate community composition relative to in-stream
reference conditions. However, catchment-wide restora-
tions are to be preferred since they may provide greater
improvement in biota beyond in-stream conditions (Haase
et al., 2013).
5 Conclusions
Re-meandering of rivers is a commonly applied measure
in river restoration with the aim to improve hydromorphol-
ogy and biotic communities. Its success largely depends
on correct implementation. Our study shows that re-
meandering should focus on the basic hydromorphological
instream requirements to improve macroinvertebrate
communities. Full reconnection of cut meanders is
capable of restoring the flow velocity, habitat composition
and macroinvertebrate communities. In contrast, recon-
nection of cut meanders with stream channels by bypass
installations does not lead to sufficient improvement of
habitats typical of flowing waters. Consequently, partial
reconnection of meanders does not suffice to significantly
improve macroinvertebrate communities and is therefore
not recommendable in environmental management.
This study was funded by the Ministry of Environment
(Landesumweltamt Brandenburg) of the federal state of
Brandenburg. We thank Uta Gr€
unert and Frank Fredrich
for their assistance during field samplings and we thank
Anne Mette Poulsen from the Department of Bioscience,
Aarhus University, for linguistic assistance. S. Lorenz was
financially supported by the European Union 7th Frame-
work Project REFORM under contract no. 282656. D.
Graeber was financially supported by the European
Union 7th Framework Project MARS under contract no.
603378.
The authors have declared no conflict of interest.
6 References
Anderson, M. 2001: A new method for non-parametric multivari-
ate analysis of variance. Austral Ecol 26,32–46.
Anderson, M. J., Willis, T. J. 2003: Canonical analysis of principal
coordinates: A useful method of constrained ordination for
ecology. Ecology 84, 511–525.
Bernhardt, E. S., Palmer, M. 2011: River restoration: The fuzzy
logic of repairing reaches to reverse catchment scale
degradation. Ecol Appl 21, 1926–1931.
Biggs, J., Corfield, A., Gron, P., Hansen, H. O., et al. 1998:
Restoration of the rivers Brede, Cole and Skerne: A joint
Danish and British EU-LIFE demonstration project, V –Short-
term impacts on the conservation value of aquatic macro-
invertebrate and macrophyte assemblages. Aquat Conserv 8,
241–255.
Bradbeer, P. A., Savage, A. A. 1980: Some observations on the
distribution and life history of Caenis robusta Eaton (Ephem-
eroptera) in Cheshire and North Shropshire. Engl Hydro-
biologia 68,87–90.
Brookes, A., Shields, F. D., Jr. 1996: River Channel Restoration.
Wiley, Chichester.
Buss, D., Baptista, D., Nessimian, J., Egler, M. 2004: Substrate
specificity, environmental degradation and disturbance struc-
turing macroinvertebrate assemblages in neotropical streams.
Hydrobiologia 518, 179–188.
Cardinale, B. J., Palmer, M. A. 2002: Disturbance moderates
biodiversity-ecosystem function relationships: Experimental
evidence from caddisflies in stream mesocosms. Ecology 83,
1915–1927.
Cardinale, B. J., Palmer, M. A., Ives, A. R., Brooks, S. S. 2005:
Diversity-productivity relationships in streams vary as a
function of the natural disturbance regime. Ecology 86,
716–726.
Cederholm, C. J., Bilby, R. E., Bisson, P. A., Bumstead, T. W.,
et al. 1997: Response of juvenile coho salmon and steelhead
to placement of large woody debris in a coastal Washington
stream. N Am J Fish Manage 17, 947–963.
Chester, E. T., Robson, B. J. 2011: Drought refuges, spatial scale
and recolonisation by invertebrates in non-perennial streams.
Freshwater Biol 56, 2094–2104.
Duan, X., Wang, Z., Xu, M., Zhang, K. 2009: Effect of streambed
sediment on benthic ecology. Int J Sediment Res 24,
325–338.
EEA [European Environmental Agency]. 2012: European waters
–Assessment of status and pressures. European Environ-
mental Agency Report No 8/2012.
Elosegi, A., Sabater, S. 2013: Effects of hydromorphological
impacts on river ecosystem functioning: A review and
suggestions for assessing ecological impacts. Hydrobiologia
712, 129–143.
Feh
er, J., G
asp
ar, J., Szurdin
e-Veres, K., Kiss, A., et al. 2012:
Hydromorphological alterations and pressures in European
rivers, lakes, transitional and coastal waters. Thematic
assessment for EEA Water 2012 Report. European Topic
Centre on Inland, Coastal and Marine Waters, Prague, ETC/
ICM Technical Report 2/2012.
Garcia, X.-F., Schnauder, I., Pusch, M. T. 2012: Complex
hydromorphology of meanders can support benthic inverte-
brate diversity in rivers. Hydrobiologia 685,49–68.
Grabowski, M., Karolina, B., Konopacka, A. 2007: How to be an
invasive gammarid (Amphipoda: Gammaroidea) compari-
son of life history traits. Hydrobiologia 590,75–84.
Graeber, D., Pusch, M., Lorenz, S., Brauns, M. 2013: Cascading
effects of flow reduction on the benthic invertebrate commu-
nity in a lowland river. Hydrobiologia 717, 147–159.
Growns, I. O., Davis, J. A. 1994: Longitudinal changes in near-bed
flows and macroinvertebrate communities in a Western
Australian stream. J N Am Benthol Soc 13, 417–438.
Gr€
unert, U., Hilt, S., Pusch, M., Gelbrecht, J. 2007: Entwicklung-
spotential der Makrophytenvegetation in der Unteren Spree
nach Renaturierungsmaßnahmen. Naturschutz und Land-
schaftspflege in Brandenburg 16,41–47.
Haase, P., Hering, D., J€
ahnig, S. C., Lorenz, A. W., Sundermann,
A. 2013: The impact of hydromorphological restoration on river
S. Lorenz et al.International Review of Hydrobiology 2016, 101,1–9
8 © 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
ecological status: A comparison of fish, benthic invertebrates,
and macrophytes. Hydrobiologia 704, 475–488.
Hamilton, J. B. 1989: Response of juvenile steelhead to in-stream
deflectors in a high gradient stream. in: Gresswell, R. E.,
Barton, B. A., Kershner, J. L. (eds.), Practical Approaches to
Riparian Resource Management. U.S. Bureau of Land
Management, Billings pp. 149–158.
Hille, S., Kristensen, E. A., Graeber, D., Riis, T., et al. 2014: Fast
reaction of macroinvertebrate communities to stagnation and
drought in streams with contrasting nutrient availability.
Freshwater Sci 33, 847–859.
J€
ahnig, S. C., Brabec, K., Buffagni, A., Erba, S., et al. 2010:
A comparative analysis of restoration measures and their
effects on hydromorphology and benthic invertebrates in 26
central and southern European rivers. J Appl Ecol 47, 671–
680.
K€
ohler, J., Hoeg, S. 2000: Phytoplankton selection in a river-lake
system during two decades of changing nutrient supply.
Hydrobiologia 424,13–24.
Laasonen, P., Muotka, T., Kivij€
arvi, I. 1998: Recovery of
macroinvertebrate communities from stream habitat restora-
tion. Aquat Conserv 8, 101–113.
Lake, P. S., Bond, N., Reich, P. 2007: Linking ecological theory
with stream restoration. Freshwater Biol 52, 597–615.
Legendre, P., Oksanen, J., ter Braak, C. 2011: Testing the
significance of canonical axes in redundancy analysis. Meth
Ecol Evo 2, 269–277.
Lepori, F., Palm, D., Br€
ann€
as, E., Malmqvist, B. 2005: Does
restoration of structural heterogeneity in streams enhance
fish and macroinvertebrate diversity? Ecol Appl 15, 2060–
2071.
Lorenz, S., Martinez-Fern
andez, V., Alonso, C., Mosselman, E.,
et al. 2016: Fuzzy cognitive mapping for predicting hydro-
morphological responses to multiple pressures in rivers.
J Appl Ecol 53 (2), 559–566.
Lowell, R. B., Culp, J. M. 1999: Cumulative effects of multiple
effluent and low dissolved oxygen stressors on mayflies at
cold temperatures. Can J Fish Aquat Sci 56, 1624–1630.
Matthaei, C. D., Werthmuller, D., Frutiger, A. 1997: Invertebrate
recovery from a bed-moving spate: The role of drift versus
movements inside or over the substratum. Archiv f€
ur Hydro-
biologie 140, 221–235.
Nagell, B., Fagerstr€
om, T. 1978: Adaptations and resistance to
anoxia in Cloeon dipterum (Ephemeroptera) and Nemoura
cinerea (Plecoptera). Oikos 30,95–99.
Noordhuis, R., van Schie, J., Jaarsma, N. 2009: Colonization
patterns and impacts of the invasive amphipods Chelicor-
ophium curvispinum and Dikerogammarus villosus in the
IJsselmeer area. The Netherlands. Biological Invasions 11,
2067–2084.
Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P., et al.
2015: vegan: Community Ecology Package. R package
version 2.3-0. http://CRAN.R-project.org/package¼vegan.
Paillex, A., Dol
edec, S., Castella, E., M
erigoux, S. 2009: Large
river floodplain restoration: Predicting species richness and
trait responses to the restoration of hydrological connectivity.
J. Appl. Ecol.46, 250–258.
Palmer, M. A., Menninger, H. L., Bernhardt, E. 2010: River
restoration, habitat heterogeneity and biodiversity: A failure of
theory or practice? Freshwater Biol 55, 205–222.
Pan, B. Z., Wang, Z. Y., Xu, M. Z. 2012: Macroinvertebrates in
abandoned channels: Assemblage characteristics and their
indications for channel management. River Res Appl 28,
1149–1160.
Pusch, M., Hoffmann, A. 2000: Conservation concept for a river
ecosystem (River Spree, Germany) impacted by flow
abstraction in a large post-mining area. Landscape Urban
Plan 51, 165–176.
R Core Team. 2014: R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna,
Austria. URL http://www.R-project.org/.
Reusch, H., Brinkmann, R. 1998: Contribution to the presence of
Trichopters species in limnetic biotopes in the lowland of North
Germany. Lauterbornia 34,91–103.
Roni, P., Bennett, T., Morley, S., Pess, G. R., et al. 2006:
Rehabilitation of bedrock stream channels: The effects of
boulder weir placement on aquatic habitat and biota. River
Res Appl 22, 967–980.
Roni, P., Hanson, K., Beechie, T. 2008: Global review of the
physical and biological effectiveness of stream habitat
rehabilitation techniques. North American Journal of Fisheries
Management 28, 856–890.
Schmedtje, U., Colling, M. 1996:
€
Okologische Typisierung der
aquatischen Makrofauna. Informationsberichte des Bayri-
schen Landesamtes f€
ur Wasserwirtschaft 4,1–548.
Schulz, M., Rinke, K., K€
ohler, J. 2003: A combined approach of
photogrammetrical methods and field studies to determine
nutrient retention by submerged macrophytes in running
waters. Aquat Bot 76,17–29.
Shields, F. D., Jr, Knight, S. S., Cooper, C. M. 2007: Can warmwater
streams be rehabilitated using watershed-scale standard
erosion control measures alone? Environ Manag 40,62–79.
Timm, H., Mardi, K.,M€
ols, T. 2008: Macroinvertebrates in Estonian
streams: The effects of habitat, season, and sampling effort on
some common metricsof biological quality. Estonian J Ecol 57,
37–57.
Usseglio-Polatera, P. 1994: Theoretical habitat templets, species
traits, and species richness: Aquatic insects in the Upper
Rh^
one River and its floodplain. Freshwater Biol 31, 417–437.
USEPA [United States Environmental Protection Agency]. 2009:
National Water Quality Inventory: Report to Congress. 2004
Reporting Cycle. United States Environmental Protection
Agency, Office of Water, Washington, DC 20460.
Van den Brink, P. J., ter Braak, C. J. F. 1999: Principal response
curves: Analysis of time-dependent multivariate responses of
biological community to stress. Environ Toxicol Chem 18,
138–148.
Verdonschot, P. F. M., Spears, B. M., Feld, C. K., Brucet, S., et al.
2013: A comparative review of recovery processes in rivers,
lakes, estuarine and coastal waters. Hydrobiologia 704,
453–474.
Wolter, C. 2010: Functional vs scenic restoration challenges to
improve fish and fisheries in urban waters. Fish Manag Ecol
17, 176–185.
International Review of Hydrobiology 2016, 101,1–9 Meander reconnection method
© 2016 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 9