Ecological Engineering 158 (2020) 106040

Contents lists available at ScienceDirect

Ecological Engineering
journal homepage: www.elsevier.com/locate/ecoleng

The impact of bioengineering techniques for riverbank protection on T

ecosystem services of riparian zones
Lars Symmanka, , Stephanie Nathob, Mathias Scholzc, Uwe Schrödera, Katharina Raupachd,
⁎

Christiane Schulz-Zunkelc
a
German Federal Institute of Hydrology (BFG), Am Mainzer Tor 1, 56068 Koblenz, Germany
b
University of Potsdam (UP), Institute of Environmental Science and Geography, Karl-Liebknecht-Str. 24-25, 14476 Potsdam-Golm, Germany
c
Helmholtz Centre for Environmental Research (UFZ), Department of Conservation Biology, Permoserstraße 15, 04318 Leipzig, Germany
d
Georg-August University Göttingen, Department of Agricultural Economics and Rural Development, Platz der Göttinger Sieben 5, 37073 Göttingen, Germany

ARTICLE INFO ABSTRACT

Keywords: Rivers and floodplains are hotspots of biodiversity and provide an exceptional number of ecological functions.
Biotechnical streambank stabilization However, they are negatively affected by human impact worldwide. The need for restoration is high, but its
Riverbank restoration realization is often hampered by antagonistic human interests. Replacing artificial riverbank protection with
Self-purification bioengineering techniques can be a first and straightforward step to restore riparian ecosystems. However,
Carbon sequestration
bioengineering still plays a marginal role in river management. We aim to introduce new arguments for
Nature-based solution
bioengineering along riverbanks by applying the ecosystem service approach. We focus on major regulating
Buffer zone
services usually provided by floodplains. Denitrification and phosphorous retention were estimated by applying
proxy-based models. Carbon sequestration within vegetation was calculated using biomass equations. Our study
clearly indicates an increase of ecosystem services by bioengineering measures compared to conventionally fixed
riverbanks. The dismantling of riprap removes up to 30 times more nitrogen and 20 times more phosphorous
from the river load. Additional slope lowering increases both values up to 50-fold. The carbon storage capacity is
four times higher in reed beds and 30 times higher in willow-brush mattresses. Our results show that bioengi­
neering techniques for riverbank protection can be a feasible tool to support general efforts towards enhancing
the self-purification of rivers and contribute to mitigating climate change, especially if conducted on a larger
scale.

1. Introduction bodies including “riparian zones” to achieve an overall “good ecological

status” of European rivers. In this context, the term “riparian” describes
Rivers and floodplains have been negatively affected by humans for semi-terrestrial areas, between the river water and the uplands which
centuries (e.g., Nilsson et al., 2005; Kingsford et al., 2016). Intensive are regularly influenced by flooding (Naiman et al., 2010). Despite
land use and the canalization of streams have altered them tre­ respectable political intentions, restoration sites are rare and ecological
mendously (Sala et al., 2000; Tockner and Stanford, 2002). In their projects are often in direct competition with antagonistic human in­
natural state, floodplains are centers of biodiversity (Naiman et al., terests (e.g. Hein et al., 2016). With a continual increase in human
1993; Catterall et al., 2007), populated by more plant and animal population, this contradiction will further intensify (Nilsson and
species than most of the other landscapes (Nilsson and Svedmark, 2002; Svedmark, 2002; MEA, 2005). Compared to other floodplain areas,
Tockner and Stanford, 2002). Apart from this, they serve as migration riverbanks are of relatively minor economic interest as they constitute
corridors enabling the genetic exchange of species over long distances only the narrow slopes between the river bed and the horizontal
(Robinson et al., 2002; Naiman et al., 2010). The ecological value of floodplain areas (Florsheim et al., 2008). Nevertheless, riverbanks in
floodplains is increasingly recognized and thus, riparian restoration has industrial countries are often modified and obstructed by artificial bank
gained importance during the past decades (e.g. Nakamura et al., 2006; protection measures to ensure safety of navigation and protect human
Feld et al., 2011). For instance, large policy guidelines like the EU- investments along the rivers (e.g. Wolter, 2001; Fischenich, 2003). In
Water Framework Directive encourage the improvement of water the German waterway network, approximately 80% of riverbanks

⁎
Corresponding author.
E-mail address: [email protected] (L. Symmank).

https://doi.org/10.1016/j.ecoleng.2020.106040
Received 24 April 2019; Received in revised form 25 June 2020; Accepted 1 September 2020
Available online 17 September 2020
0925-8574/ © 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

recorded in management plans are obstructed by artificial bank pro­ retention in a test track of BRP in Germany. Additionally, carbon se­
tections (L. Symmank, unpublished data). These human induced mod­ questration in biomass of bank vegetation was calculated applying al­
ifications have led to a unification of bank structures (e.g. substrates lometric equations. To our knowledge our study is a first approach to
and bank inclination) and thus a loss of habitat diversity (Florsheim analyze ecological values of BRP beyond biodiversity. We hypothesize
et al., 2008). As a consequence, typical wetland species decline in favor that the implementation of BRP leads to a significant increase in se­
of terrestrial species (Harvolk et al., 2015; Wollny et al., 2019). From an lected ecosystem services. On the example of the German waterways,
ecological point of view, dynamic and unspoiled riverbanks are essen­ we aim to show that the application of BRP on a larger scale can con­
tial for river sustainability (Florsheim et al., 2008). However, as long as tribute to general efforts to increase water quality and mitigate climate
stability aspects have priority in river management, humans will try to change. Our results illustrate additional ecological benefits of BRP and
impede the natural dynamic of riverbanks. its potential application to help decision makers facing the enormous
Bioengineering techniques for riverbank protection (BRP) are pro­ ecological challenges in future river management.
mising tools to promote natural conditions on watercourses where
erosion control is needed (Janssen et al., 2019). Thereby, conventional 2. Methods
bank fixation (e.g. riprap) is replaced by biological components that
undertake bank stabilization (Li and Eddleman, 2002, Wells 2002). 2.1. Bank types
Roots of native trees (Evette et al., 2012) or herbs (e.g. Pinto et al.,
2016) assume the role of hard engineering and stabilize riverbanks Our study is based on data from a test track of BRP at the River
sufficiently (Norris et al., 2008). Depending on site conditions, BRP can Weser in north-western Germany (52°30′N; 09°04′E). The moderately
also be associated with civil engineering (e.g. Wu and Feng, 2006; navigated river belongs to the German waterway network. In our study
Frothingham, 2008). In any case, BRP create habitats that provide more area, the Weser is a typical low-land watercourse with a comparatively
niches and resources than conventional bank fixation. An increasing large catchment and a floodplain dominated by agriculture (Brunotte
number of studies illustrates the positive effects of BRP on biodiversity et al., 2009). The width of the river at the study area is approximately
(e.g. Li et al., 2006, Cavaillé et al., 2015; Schmitt et al., 2018, Tisserant 80 m with a mean discharge of 180 m3/s (BfG and BAW, 2008). Like
et al., 2020). Although the use of BRP has increased during the last few other German waterways, large parts of the watercourse are im­
decades (e.g. Wu and Feng, 2006; Evette et al., 2009; Anstead and Boar, pounded, leading to a decrease of water-level fluctuations (Scholz et al.,
2010), they are still in limited quantities affecting only a small fraction 2012). The riverbanks are characterized predominantly by riprap de­
of human-impacted riverbanks. This may be because ecological aspects posited to avoid bank erosion. The test track was installed in the winter
are still overlooked in decision making in favor of alleged financial of 1989 and consists of different combinations of BRP (BfG and BAW,
perspectives (Fischenich and Allen, 2000; Evette et al., 2009) or prac­ 2008). Two test fields within the test track proved to be particularly
titioners may simply not be aware of the long-term benefits of BRP (e.g. successful regarding bank stability and thus were chosen for further
Jabłońska et al., 2020). investigations. The selected BRP test fields are compared to a con­
We aim to introduce new arguments for the implementation of BRP ventionally fixed riverbank (hereafter named: bank type “riprap”). All
by applying the ecosystem service approach (MEA, 2005). Worldwide, investigated bank types had a length of 30 m and a width of 5 m cor­
ecosystems deliver a wide range of functions that are directly or in­ responding to a ground area of 150 m2. The bank type “riprap” has a
directly used as ecosystem services for human well-being (Costanza slope inclination of approximately 20 degrees with a prominent riprap
et al., 1997). Floodplains deliver an exceptional number of ecosystem cover of 0.6 m thickness (Fig. 1).
services. The Millennium Ecosystem Assessment (MEA, 2005) lists 17 In the first selected BRP test field (hereafter named: bank type
different provisioning, regulating, cultural and supporting services of “softwood”), riprap was removed to 0.4 m below the mean water level
floodplains, with particularly high magnitudes for food production, (Fig. 1). Willow brush mattresses consisting of Salix viminalis L. were
fiber and fuel, climate regulation, soil formation and nutrient cycling. installed to prevent bank erosion. After 28 years of undisturbed
In general, the overall benefits of wetlands, including floodplains, are growing, the brush mattresses developed into a dense softwood coppice
higher than those of other terrestrial ecosystems (McClain et al., 2003, (Fig. 2). In the second selected BRP test field (hereafter named: bank
Vidon, 2010, Acreman et al., 2011). type “reed”); the riprap was also removed and additionally the slope
In our study, we focus on major regulating services usually provided inclination was reduced to less than 10 degrees (Fig. 1). Initial plantings
by natural floodplains. The removal or retention of nutrients such as with common reed (Phragmites australis [Cav.] Steud.) were carried out
nitrogen (N) and phosphorous (P) from surface water has been re­ and developed into a densely growing reed bed (Fig. 2).
peatedly highlighted (e.g. Olde Venterink et al., 2006; Newcomer
Johnson et al., 2016). The river ecosystems are capable of self-pur­ 2.2. Nutrient retention
ification and reduce nutrients through bacterial decomposition or fil­
tration (e.g. Brettar et al., 2002; Hoffmann et al., 2009). Furthermore, The removal of nitrogen from river water is mainly based on deni­
they are considered to be sinks of greenhouse gasses (Sutfin et al., trification processes within floodplain areas during flood events (Pinay
2016), sequestering organic carbon in particulate or dissolved condi­ et al., 2002; Verhoeven et al., 2006). Thereby, nitrate from surface
tions in soils and in biomass (Kayranli et al., 2010). water is converted into atmospheric nitrogen (Trepel and Palmeri,
We apply proxy-based models to estimate the N-removal and P- 2002). The process is fostered by anaerobic soil conditions and elevated

Fig. 1. Transverse geometry of the examined riverbank types derived from a technical report (BfG and BAW, 2008). Type “riprap”: riprap (stained) with natural soil
substrate below (grey), bank inclination 20 degrees. Type “softwood”: riprap removed below mean water level, bank inclination 20 degrees. Type “reed”: riprap
removed, bank inclination 10 degrees. The aquatic (blue) and terrestrial (orange) study sections and respective dimensions are illustrated below the profile (A-E).
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

2
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

Fig. 2. Measure types investigated at the test track on the River Weser. Bank type “riprap” (conventional riverbank in vicinity of the test track, 2019, photo: E.
Kramer), bank type “softwood” (2015), bank type “reed” (2015).

C-content within floodplain soils (DeLaune et al., 1996; Waters et al., measures were attributed to the land-use class “water”. With regard to
2014). Phosphorous in rivers is mainly bound to particles (Olde P-retention both sections were assigned to the “shore area”. The ter­
Venterink et al., 2006). Thus, P-retention is mainly determined by se­ restrial section C consists of a willow brush mattress and was assigned
dimentation processes within floodplain areas during flood events to the land-use class “forest”. Section E was assigned to the land-use
(Hoffmann et al., 2009). Surface roughness from vegetation and the class “wetland”, since slope lowering increases inundation frequency
duration of floods are decisive factors for sedimentation (Gautier et al., and reed development is high in this area (Schmitt et al., 2018).
2009). The nutrient loads of rivers in developed countries are usually
high due to exalted N and P input from urban and agricultural areas 2.2.2. Denitrification
(Allan, 2004; Dodds, 2006; Feld et al., 2011). Thus, the nutrient load of The assignment of land-use classes to denitrification levels follows
the river is not expected to be a limiting factor here. For simplicity, we the approach described in Natho et al. (2013) illustrated in Fig. 3.
use the term ‘nutrient retention’ in the following text when talking According to the authors non-vegetated surfaces are not considered
about N and P in their entirety even though we are aware that deni­ relevant for denitrification. However, matter exchange between water
trification removes nitrogen from the system. and floodplain soil cannot be excluded in riprap and few existing plants
We used two proxy-based models to estimate nutrient retention do create a small amount of autochthonic organic soil in the gaps be­
rates for specific bank sections identified in Fig. 1. The approaches are tween the rocks. Thus, we choose the lowest denitrification level 1
based on extensive literature review of various European field mea­ (5 kg N ha−1 yr−1) for section A even if we have likely overestimated
surements to infer universally applicable proxy values for nutrient re­ the actual denitrification potential of bank type “riprap”. The aquatic
tention in floodplains (Schulz-Zunkel et al., 2012; Natho et al., 2013). sections B and D of bank types “softwood” and “reed” provide an es­
Two decision cascades are presented for N and P (Figs. 3 and 4). Factors timated denitrification rate of 300 kg N ha−1 yr−1 (level 6). The crucial
influencing the models are land-use type, soil type, floodplain condition factor for the assignment of denitrification rates within forests is the soil
and surface roughness. As a result, different levels of nutrient retention type. According to the soil map BUEK50 (NIBIS®, 2014), the soil type in
were estimated. All values were given in kilograms of reduced N or P terrestrial section C of bank type “softwood” is vega (comparable to
per hectare per year (kg ha−1 yr−1). Due to the direct contact of riv­ fluvisol), leading to denitrification level 3. However, if the floodplain
erbanks with the water body, all studied bank types can be classified as condition is “moderately modified” or better according to the defini­
active floodplain for which the decision cascades were initially devel­ tions of Brunotte et al., 2009 (Follner et al., 2010) the final deni­
oped. Retention levels were assigned to bank sections as follows: trification rate can be elevated by one level. Due to the close proximity
to the river, the associated high flooding potential and the low land-use
2.2.1. Land-use classes intensity, bank type “softwood” at least fulfills the definition of “mod­
The assignment of land-use classes to the sections according to erately modified”. Thus, we assigned an average denitrification rate of
Fig. 1 is essential for both retention models and was carried out ac­ 100 kg N ha−1 yr−1 (level 4) to section C. The terrestrial section E of
cording to the vegetation found in the respective bank types. The ter­ the bank type “reed “has an estimated denitrification rate of
restrial riprap of section A consists of rocks, artificial geotextile beneath 250 kg N ha−1 yr−1 (level 5).
it and only a few plants growing in the gaps between the rocks. Thus,
this bank type can generally be considered as “non-vegetated” habitat. 2.2.3. P-retention
In the case of denitrification, the aquatic sections B and D of the BRP The assignment of P-retention levels follows the recommendations

Fig. 3. Decision cascade for assessing denitrification rates in floodplains (simplified from Natho et al., 2013).

3
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

Fig. 4. Decision cascade for assessing phosphorous retention in floodplains and rivers (simplified from Schulz-Zunkel et al., 2012).

of Schulz-Zunkel et al. (2012) illustrated in Fig. 4. Due to the sparse equations for aboveground biomass, including stems, branches and
vegetation and the uneven structure of the riprap we assume a very low bark for Salix exigua Nutt. This North American willow species grows
surface roughness and thus apply the lowest P-retention level 5 (0.5 kg exclusively shrub-like and has a similar natural height as Salix viminalis
P ha−1 yr−1) for bank type “riprap”. While no relevant sedimentation (Collet, 2004). For comparison, we additionally estimated the stem
usually occurs in the largest part of the river, a very high P-retention biomass based on equations for arborescent species summarized in
was measured in aquatic shore areas (Kronvang et al., 1999). Thus, Zianis et al. (2005). The authors present three different calculations of
sections B and D of bank type “softwood” and “reed” receive a high P- stem volumes for willow species. To derive the stem biomass, we con­
retention value of 50 kg P ha−1 yr−1 (level 7). The terrestrial sections C verted the stem volume using the equation suggested by Jalkanen et al.
and E were assigned to retention level 1 for forest and wetland re­ (2005). An age-dependent biomass expansion factor for broadleaved
spectively with a corresponding retention rate of 5 kg P ha−1 yr−1. An stands was obtained from Lehtonen et al. (2004). All input data (stem
overview of all nutrient retention rates assigned is given in Table 1. DBH), applied equations and intermediate results are summarized in
the supplementary material.Tree components missing in the approaches
of Matzek et al. (2014) and Zianis et al. (2005) were derived from the
2.3. Carbon sequestration in biomass
proportions of components out of the total biomass as suggested in
Frouz et al. (2015) (see Table 2).
We estimate the additional C-storage of BRP by comparing the ve­
To estimate the C-stock in bank type “reed “we applied the biomass
getation cover for the studied bank types. C-content is derived from
values suggested by Zerbe et al. (2013) for Phragmites australis growing
estimations of the dry weight of biomass (termed only “biomass” from
in north-eastern Germany. The authors measured a mean aboveground
here on) multiplied by vegetation-specific conversion factors (see
biomass in rewetted degraded peat land of 12.5–23.8 t ha−1. Below­
below). Following Penman et al. (2003), the obtained carbon stocks are
ground biomass ranged between 1 and 2 times the amounts of above­
multiplied by the factor of 3.67 to receive the CO2 equivalents detracted
ground biomass (Zerbe et al., 2013). We applied conservative values of
from the atmosphere.
12.5 t ha−1 for aboveground and belowground biomass respectively.
Vegetation cover in bank type “riprap” usually does not exist, except
These values closely match measurements published in Allirand and
for some sporadically growing herbaceous perennials in the gaps be­
Gosse (1995) from central France. The biomass was multiplied by 0.45
tween the rocks. The overall biomass for this bank type was negligible.
according to Zerbe et al. (2013) to obtain the total C-content of reed
However, we added a hypothetical value of 0.5 kg m−2 for further
plants.
calculation even if we likely overestimate the actual amount of biomass
within this bank type. According to Alexeyev and Birdsey (1998) the
proportion of carbon within dry biomass is around 50%. Thus, we 3. Results
generally multiplied the biomass by a factor of 0.5 to derive the C-
content of the plants. 3.1. Nutrient retention
The brush mattresses of bank type “softwood” consist of densely
growing willows (mainly Salix viminalis L.) that are 7 m high on Both BRP bank types show higher nutrient retention values than
average. The shrubs shade out herbaceous vegetation and constitute the bank type “riprap”. This is mainly due to the removal of riprap enables
only relevant vegetation structure in this bank type. We used allometric direct contact between the river and natural floodplain soil where de­
equations of several studies to estimate the woody biomass of the wil­ nitrification takes place. Furthermore, a natural river bank vegetation
lows. This gentle method is widely applied (e.g. Cierjacks et al., 2010; develops which increases sedimentation processes. The river expands
Tigges et al., 2017) and minimizes interventions into rare ecosystems. inland, leading to additional aquatic areas with increased nutrient re­
The biomass is estimated by measuring the stem diameter at breast tention. The additional slope lowering in the bank type “reed” leads to
height (DBH, 1.3 m above ground) of single stems along with the height the highest retention rates of all three analyzed bank types. The results
of plants. In our study, DBH values ranged from 2 to 18 cm. In total, 36 are summarized in Table 1.
individuals (polycorms) containing 314 single stems were measured.
Despite an extensive literature review, specific equations for Salix 3.1.1. Denitrification
viminalis could not be found. However, we found a few equations for Compared to the BRP bank types, the denitrification potential of
closely related Salix species. Considering the poor data situation, we bank type “riprap” is low and comprises estimated 0.075 kg N yr−1. The
applied all available equations and discuss the different results below. denitrification rate of bank type “softwood” is substantially higher. The
This approach gives us a first estimation of possible carbon sequestra­ removal of riprap expands the area below the mean water level by more
tion in willow brush mattresses. Frouz et al. (2015) developed an than one meter (section B, Fig. 1). Compared to the previous terrestrial
equation for Salix caprea L. which usually grows in arborescent (tree- riprap, the denitrification rate increases by a factor of 60. The terrestrial
like) form. However, the authors developed a particular equation for section C is transformed from an artificial bank stabilization into a
shrub-like individuals on post-mining sites. Matzek et al. (2014) applied natural floodplain habitat (softwood). The floodplain status of this

4
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

Table 1
Overview of essential parameters, applied model rates and the resulting absolute values of denitrification and phosphorous retention for the riverbank types studied
(figures not drawn to scale).

bank type riprap softwood reed

section A B C D E

area in m2 150 36 114 84 66

land-use class non- wetland

river forest river
vegetated
soil typea none/
vega vega vega vega
geotextile
floodplain heavily moderately moderately
Denitrification

- -
statusb modified modified modified
denitrification 5 300 100 300 250
rate (level 1) (level 6) (level 4) (level 6) (level 5)
(kg ha-1 yr-1)c
1.08 1.14 2.52 1.65
denitrification
0.08
(kg yr-1) 2.22 4.17

surface high
very low - high -
roughnessd
P-retention 0.5 50 5 50 5
P-retention

rate (level 5) (level 7) (level 1) (level 7) (level 1)

(kg ha-1 yr-1)d
0.180 0.057 0.42 0.03
P-retention
0.01
(kg yr-1) 0.24 0.45

Bank type Riprap Softwood Reed

Section A B C D E

Area in m2 150 36 114 84 66

Denitrification Land-use class Non-vegetated River Forest River Wetland

Soil typea None/geotextile Vega Vega Vega Vega
Floodplain statusb Heavily modified – Moderately modified – Moderately modified
Denitrification rate (kg ha−1 yr−1)c 5 (level 1) 300 (level 6) 100 (level 4) 300 (level 6) 250 (level 5)
Denitrification (kg yr−1) 0.08 1.08 1.14 2.52 1.65
2.22 4.17
P-retention Surface roughnessd Very low – High – High
P-retention rate (kg ha−1 yr−1)d 0.5 (level 5) 50 (level 7) 5 (level 1) 50 (level 7) 5 (level 1)
P-retention (kg yr−1) 0.01 0.180 0.057 0.42 0.03
0.24 0.45

a
Soil types derive from the soil map BUEK50 (NIBIS®, 2014).
b
Based on Brunotte et al., 2009.
c
Described in Natho et al., 2013.
d
Based on Schulz-Zunkel et al. (2012).

section improves considerably, leading to an increase in the deni­ The denitrification rate of the terrestrial section is 50 times higher than
trification rate by a factor of 20. The overall gain in denitrification for riprap. The overall gain of N-removal in bank type “reed “is almost
compared to the conventional riverbank (“riprap“) in bank type “soft­ twice as high as in bank type “softwood “and accounts for more than
wood” is more than 2 kg yr−1. In bank type “reed”, the additional 4 kg N yr−1.
lowering of the slope expands the aquatic section D to 2.8 m (Fig. 1).

5
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

Table 2
Calculated and derived biomass values (BM) of surveyed willow plant parts.
Literature source Stem Vol. (m3) Stem BM (t) Branch BM (t) Dead wood BM (t) Leaf BM (t) Coarse root BM (t) Total BM (t) C stock (t)

Proportion on total biomass* 37.4% 9.8% 15.2% 12.2% 25.4% 100%

Frouz et al. (2015) 2.52 1.26
Matzek et al. (2014) 1.88 0.37 0.77 3.02 1.51
Zianis et al. (2005) -219 4.61 2.52 0.66 1.02 0.82 1.71 6.73 3.37
Zianis et al. (2005) -220 3.12 1.71 0.45 0.70 0.56 1.16 4.58 2.29
Zianis et al. (2005) -221 3.02 1.65 0.43 0.67 0.54 1.12 4.41 2.21

Values printed in bold directly derive from equations of the literature sources listed. 1-3 Stem volume calculations are based on three different equations for willow
species published in Zianis et al. (2005, Appendix A, number 219-221). Normal style values are derived from proportions of plant components according to Frouz
et al. (2015)⁎. The carbon content of woody plants is half of the biomass according to Alexeyev and Birdsey (1998).

3.1.2. P-retention et al., 1997; Schultz et al., 2005). Furthermore, in their natural state,
Analogous to denitrification, the P-retention in bank type “riprap” is they can also store large amounts of carbon within biomass (Fortier
low. The comparatively smooth structure of rocks only allow a limited et al., 2015). In this context, our results clearly illustrate the ecological
sedimentation trapping efficiency and thus only a negligibly small benefits of BRP compared to conventionally fixed riverbanks. Large
amount of annual P-retention (< 0.01 kg P yr−1). In the aquatic section amounts of N and P are retained due to riprap removal and slope
B of bank type “softwood”, the P-retention rate is a hundredfold higher lowering. The introduction of living plant material led to a considerable
than in terrestrial riprap. The subsequent shrubby vegetation in section increase in C-sequestration and thus additional capture of CO2. Al­
C leads to higher surface roughness and thus multiplies the retention though the overall ecological advantages of BRP are significant the
rate in this area by a factor of 10. The overall gain in retained P in bank quantities of the respective ecosystem services are unequal among the
type “softwood” is 0.23 kg yr−1. Bank type “reed “yields the highest P- different measure types.
retention rates of all bank types studied, since the size of the aquatic
section D is larger than in the bank type „”softwood“. The net gain in 4.1. Denitrification
retained P accounts for more than 0.44 kg P yr−1.
The removal of riprap in BRP measures has a fundamental impact on
3.2. Carbon sequestration in biomass denitrification. On the one hand it enables direct contact between the
river water and the floodplain soil. On the other hand, the riverbanks
The hypothetical biomass value of 0.5 kg m−2 on riprap results in a are lowered by the width of the previous riprap layer. This increases the
total biomass of 75 kg in bank type “riprap”, which corresponds to flooding frequency and thus anaerobic soil conditions, which positively
37.5 kg of carbon and 138 kg of CO2. This amount is significantly lower impacts denitrification (Burt et al., 1999; Natho et al., 2013). The latter
than the biomass of the BRP bank types studied. The woody vegetation also benefits from increased organic carbon input from litter fall and
of the bank type “softwood” accumulates large amounts of above- and decaying root biomass of the willow-brush mattress in bank type
belowground biomass. However, the results based on different allo­ “softwood “(Burt et al., 1999; Brettar et al., 2002; Ghestem et al., 2011).
metric equations vary considerably (Table 2). The equations of the All this increases the overall denitrification in bank type “softwood” by
shrub-like willow species Salix caprea (Frouz et al., 2015) and Salix almost a factor of 30 (Table 1). The positive effect of vegetation on
exigua (Matzek et al., 2014) result in a significantly lower biomass than denitrification was also shown by Sweeney et al. (2004) and Hefting
those run for solitary trees (Zianis et al., 2005). Based on different et al. (2005), who compared forested and deforested river reaches.
biomass calculations, the derived carbon content of willows in this bank Kaushal et al. (2008) found similar patterns in restored and unrestored
type ranges between 1.26 and 3.37 tons. The resulting retention of CO2 reaches of urbanized watersheds.
is 30 to 90 times higher than in the bank type “riprap” (Table 3). The A large input of organic carbon can also be assumed from the annual
value of 12.5 t ha−1 for the aboveground biomass of reeds and the decay of aboveground plant parts of Phragmites australis in the bank
assumption of a 1:1 ratio of stem and root biomass results in a total type “reed”. The additional flattening of the slope considerably in­
biomass of 375 kg for bank type “reed“. This corresponds to 169 kg of creases flooding frequency. More than half of the area is permanently
total carbon stored within above and belowground plant parts. This inundated, because larger parts of the riverbanks are below the mean
means that more than 620 kg of CO2 can be stored, which is con­ water level. The remaining area is frequently inundated even during
siderably less than the bank type “softwood” but more than four times smaller flood events which should result in higher terrestrial deni­
the amount of the bank type “riprap”. trification rates. Thus, the overall denitrification in bank type “reed”
increases by almost a factor of 50 (Table 1).
4. Discussion According to our findings, one kilometer of the bank type “reed
“yields a mean turnover of about 139 kg N yr−1. Large-scale floodplain
Vegetated riverbanks are highly valuable, as they are able to remove restoration projects provide much higher metabolic denitrification
large amounts of N and P from the river catchment (Jabłońska et al., rates. For example, a dike relocation along the German River Elbe
2020). They are also important buffer zones that actively hinder nu­ processes additional 96 t N yr−1 for a measured size of approx. 600 ha
trients from entering the river system from agricultural areas (Lowrance (Lautenbach et al., 2012). To retain a comparable amount of N through

Table 3
Carbon (C) content and carbon dioxide equivalents of the bank types studied referred to the ground area of 150 m2.
Bank type C-content (t) C-content (t ha−1) CO2-equivalent (t) Additional CO2(t) CO2-storage (t ha−1)

Riprap 0.04 2.67 0.15 – 9,79

Softwood 1.26–3.37 84.00–244.67 4.62–12.37 4.48–12.23 308.28–824.53
Reed 0.17 11.33 0.62 0.47 41.59

CO2 derived from C content multiplied by the factor of 3.67 according to Penman et al. (2003).

6
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

BRP, approximately 690 km of riverbanks have to be modified into the and permanent C-stock. Common reed is a densely and tall-growing
bank type “reed”. Although this figure appears to be high, it only ac­ perennial grass with bulky rhizomes and a large amount of necromass
counts for 5% in riverbanks of the German waterway network (~ (Zerbe et al., 2013). The gain in atmospheric CO2 retained in the bank
14.000 km) (BMVI, 2019). Furthermore, the denitrification rates ap­ type “reed “is approximately 0.5 tons and thus at least four times higher
plied in our study were rather conservatively chosen. A recent study in than in bank type “riprap” (Table 3). Significantly more CO2 is retained
a comparable and nearby river system demonstrated much higher de­ in the woody biomass of the bank type” softwood”, even if results based
nitrification rates within the aquatic zone (Ritz et al., 2018). The ap­ on different allometric equations vary considerably (Table 2). This is
plication of this result (approx. 1200 kg ha−1 yr−1) to the aquatic parts mainly due to the different habits of the plant species for which equa­
of bank type “reed” (section D, Fig. 1) would increase the net deni­ tions were generated. The highest biomass values derive from equations
trification rate per stream kilometer to 391 kg N yr−1. As a result, the of solitary standing trees (Zianis et al., 2005). However, the growth
number of bank kilometers needed to gain the denitrification potential form and height of willows measured in our study are more similar to
of the dike relocation reduces to only 246. This length corresponds to the shrub-like willows described by Frouz et al. (2015). Thus, and to
less than 1.7% of the German waterway network. avoid an overestimation of carbon-storage capacity in further discus­
sions we assume a low additional amount of 4.48 tons of CO2 stored in
4.2. P-retention the bank type” softwood” (Table 3). Nevertheless, the C-sequestration
in type “softwood” is at least more than seven times higher than in type
Vegetation is a crucial factor in the provision of P-retention in “reed” and more than 30 times higher than in bank type “riprap”. Our
floodplains. Densely-growing plants decrease the flow velocity during assumptions are in line with Fortier et al. (2015), who found 9–31 times
flood events and thus increase sedimentation processes (Schneider more carbon-storage capacity in forested reaches compared to their
et al., 2006; Hoffmann et al., 2009). Due to the rather smooth surface of herbaceous counterparts. Further significant amounts of carbon storage
riprap, the P-retention potential of the bank type “riprap “is low com­ can be found in fine roots, which generally increase with forest age
pared to naturally vegetated riverbanks. On the contrary, shrubby ve­ (Giese et al., 2003; Matzek et al., 2014) but have not been estimated
getation has high roughness values (Pasche et al., 2006), which results here.
in the highest terrestrial P-retention values in the willow-brush mat­ If we convert our results, approximately 168 tons of biomass per
tresses of bank type “softwood”. Here, the overall P-retention is more hectare are accumulated over 28 years of undisturbed development of
than twenty times higher than on conventionally fixed riverbanks the bank type” softwood”. The results correspond with the measure­
(Table 1).These findings are supported by several studies, such as Søvik ments from short-rotation plantations, which are treated as potential
and Syversen (2008), who reported a higher P-retention in riverbanks carbon sinks (Rytter, 2012). Measurements of willow plantations (Salix
vegetated by trees than by grass. Rieger et al. (2014) showed that viminalis) result in an aboveground biomass between 4.5 t ha−1
especially willows show significantly high sedimentation rates com­ (Gruenewald et al., 2007) and 70 t ha−1 (Hofmann-Schielle et al.,
pared to other tree species. Bank type “reed” exhibits similarly high P- 1999) after growing periods of 5–6 years. Riparian buffers with poplar
retention values. The densely growing and comparatively high reed trees (Populus spec.) have been recorded to accumulate a total biomass
plants induce an even higher surface roughness than willows (Pasche of 52 t ha−1 for 7-year old stands (Tufekcioglu et al., 2003) and 66 to
et al., 2006). Thus, with regard to P-retention, Olde Venterink et al. 223 t ha−1 for 9-year old stands (Fortier et al., 2015). Converted into
(2006) recommend focusing on reed beds when restoring floodplain carbon, approximately 84 tons per hectare are sequestered in bank type
areas. However, our applied retention model does not distinguish be­ “softwood”. Larger amounts of carbon are stored in mature riparian
tween willows and reed. The actual P-retention in bank type “reed” may forests. Rheinhardt et al. (2012) calculated an aboveground carbon
thus be even higher. At any rate, the largest increase in P-retention was stock of 242 t ha−1 in forested riparian woodland in North Carolina.
obtained by the slope lowering and related enlargement of the aquatic Similar amounts were found in the study by Fierke and Kauffman
zone in bank type “reed“. As a consequence, the overall P-retention is (2005), who measured an aboveground biomass of 437 t ha−1 (corre­
almost fifty times higher than that of conventionally fixed riverbanks. sponding to 219 t ha−1 of C) in riparian cottonwoods of Western
In terms of P-retention, slope lowering in particular makes the bank Oregon. Cierjacks et al. (2010) reported 163 t ha−1 of carbon in soft­
type “reed” the most productive measure type studied. wood of the Danubian National Park in Austria. The latter is approxi­
According to our results, one kilometer of the bank type “reed” mately twice the rate of carbon stored in willow-brush mattresses,
retains up to 15 kg P yr−1. However, maximum values of P-retention in which makes riverbank type “softwood” an attractive method to absorb
floodplains can be substantially higher (Hoffmann et al., 2009). Studies carbon dioxide in otherwise treeless floodplain areas.
from the Netherlands measured P-retention rates in riparian reed beds The absolute amount of carbon stored in softwoods inside the
of 95 kg P ha−1 yr−1 (Olde Venterink et al., 2006). Wolter (2001) Danubian National Park in Austria is about 30.000 tons (Cierjacks et al.,
concluded that the restoration of only one-fifth of German riverbanks 2010). Generally, floodplain forests store huge amounts of carbon
would result in a substantial improvement in the strongly limited fish compared to other terrestrial ecosystems (Cierjacks et al., 2010).
diversity. If that number of riverbanks along German waterways (~ However, globally such forests have largely been cleared since the
2800 km) would be restored according to the bank-type “reed”, be­ emergence of human settlements (UNEP-WCMC, 2000; Angradi et al.,
tween 42 t P yr−1 (values used in this study) and 98 t P yr−1 (values 2004; Nakamura et al., 2006). The afforestation of riverbanks is a first
examined by Olde Venterink et al., 2006) could be retained. The latter and straightforward step to increase the biomass in human modified
value is close to the overall P-retention of the entire active floodplain of floodplains. To store a similar amount of C in bank type” softwood”
the largest German River Rhine (~ 120 t yr-1; Schulz-Zunkel et al., (42 t of C per stream kilometer) as in the softwood of the Danubian
2012). National Park, approximately 714 km of riverbanks would have to be
modified, corresponding to 5.1% of banks of German waterways. This
4.3. Carbon sequestration in biomass proportion appears to be manageable, considering that the Danubian
National Park presents the largest wetland environment subjected to
Conventional riverbank fixations such as riprap store only small natural fluvial dynamics in Central Europe. The greening of riverbanks
amounts of carbon within the biomass (Table 3). With increasing age, by willow brush mattresses is thus an ideal example for the reforesta­
riprap can become overgrown with herbaceous vegetation or even carry tion of degraded areas which is one of the major strategies suggested to
patches of shrubs and small trees. However, compared to natural riv­ mitigate global climate change (Canadell and Raupach, 2008;
erbanks, the vegetation cover remains low and is often cut back during Rheinhardt et al., 2012).
maintenance work. The bank type “reed “exhibits a considerably higher

7
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

4.4. Uncertainties 2015; von Haaren et al., 2016) could supportively apply BRP to connect
natural floodplain remnants. Here, BRP can contribute to current efforts
Our study is a first estimation of the provision of regulating eco­ to support the self-purification of rivers and mitigate climate change.
system services through BRP. However, we are aware that several un­ This makes them an ideal case study to illustrate the benefits of man­
certainties remain. First of all, our assumptions regarding nutrient re­ agement changes to ecosystem functionality and human well-being.
tention are based on land-use classes on the landscape scale, with a
spatial resolution of 25 m (Schulz-Zunkel et al., 2012, Natho et al., Glossary
2013,). At first glance, the rather coarse resolution seems unsuitable for
small-scale measures like BRP. To reach this point, we have used so- Riprap A foundation of large chunks of stone thrown together irre­
called conservative values so as not to overestimate the nutrient re­ gularly or loosely on a soft bottom to protect banks from
tention potential. So, it is likely that the actual values are in part con­ erosion
siderably higher. Furthermore, the chosen method does not take into Willow-brush mattress Live cuttings of willows with branches installed
account particular chemical influencing factors on nutrient retention. parallel to the slope direction and with soil contact to form a
For instance, the river load (Walton et al., 2020) or pH level of the soil mattress
(Šimek and Cooper, 2002) may have large impacts on nutrient retention
and vary from case to case. Otherwise, our approach enables us to make Declaration of Competing Interest
general statements about nutrient retention within BRP measures in­
dependent of the study area. Regarding C-sequestration our results are None
also based on estimations instead of measurements. The latter could
only be obtained by “destructive sampling”, which requires the removal Acknowledgements
of all relevant plant parts (Nordh and Verwijst, 2004). This approach
would lead to severe interventions in our small study areas. Our results We sincerely thank Andreas Sundermeier, Peter Horchler, Helmut
give a first and considerable indication of the increased carbon se­ Fischer and Kathrin Schmitt for their constructive comments, Anne
questration in the studied BRP types without destroying valuable and Popiel, Sarah Gwillym-Margianto and Paul Hudson for proofreading the
rare habitats. As mentioned before we used the lowest results for dis­ article as well as Sarah Wagner, Björn Hoppe, Jutta Buscher and Dieter
cussion. Again, the actual values of C-sequestration might be higher. It Nelle for technical support. We also thank Uwe Rode from the
is desirable that future field measurements of ecosystem functions are Waterways and Shipping Administration for helpful information and
also conducted in BRP to confirm or disprove our results. Furthermore, support during the field work. The study was financed by the German
some aspects of ecosystem functions studied here could not be con­ Federal Ministry of Transport and Digital Infrastructure (BMVI).
sidered in our approach. These include, for example, the biomass of fine
roots or the re-mobilization of sediment. The latter can represent a Appendix A. Supplementary data
relevant re-entrance of P into watercourses (Hoffmann et al., 2009).
Furthermore, we only considered a small number of the ecosystem Supplementary data to this article can be found online at https://
services that are provided by floodplains. For example, nutrient uptake doi.org/10.1016/j.ecoleng.2020.106040.
by plant biomass (e.g. Tufekcioglu et al., 2003) or cultural benefits (e.g.
Symmank et al., 2020) are also significant services of restored riv­ References
erbanks. Last but not least, other processes such as methane flux from
wet floodplain soils (Brix et al., 2001) could counteract the ecosystem Acreman, M.C., Harding, R.J., Lloyd, C., McNamara, N.P., Mountford, J.O., Mould, D.J.,
services studied here (trade-offs). Purse, B.V., Heard, M.S., Stratford, C.J., Dury, S.J., 2011. Trade-off in ecosystem
services of the Somerset Levels and Moors wetlands. Hydrol. Sci. J. 56 (8),
1543–1565. https://doi.org/10.1080/02626667.2011.629783.
5. Conclusion Alexeyev, V.A., Birdsey, R.A., 1998. Carbon storage in forests and peatlands of Russia.
USDA For. In: Serv. Northeastern Res. Statn., Radnor, Penn. Gen. Tech. Rep. NE-244.
In their natural state, riverbanks are characterized by dynamic [Initially published in Russian in 1994].
Allan, J.D., 2004. Landscapes and riverscapes: the influence of land use on stream eco­
erosion and sedimentation. However, as long as humans prefer bank systems. Annu. Rev. Ecol. Evol. Syst. 35, 257–284. https://doi.org/10.1146/annurev.
stability rather than natural processes along riverbanks, BRP are an ecolsys.35.120202.110122.
ideal example for “nature-based solutions” (Cohen-Shacham et al., Allirand, J.M., Gosse, G., 1995. An above-ground biomass production model for a
common reed (Phragmites communis Trin.) stand. Biomass Bioenergy 9 (6),
2016) and a promising tools to increase riverbank ecosystem func­
441–448. https://doi.org/10.1016/0961-9534(95)00042-9.
tionality on intensively used rivers. Compared to conventionally fixed Angradi, T.R., Schweiger, E.W., Bolgrien, D.W., Ismert, P., Selle, T., 2004. Bank stabili­
riverbanks, BRP clearly improve the provision of selected regulating zation, riparian land use and the distribution of large woody debris in a regulated
reach of the upper Missouri River, North Dakota, USA. River Res. Appl. 20 (7),
ecosystem services. Thereby, the intensity of particular ecosystem
829–846. https://doi.org/10.1002/rra.797.
functions differs between BRP types. The removal of riprap is equally Anstead, L., Boar, R.R., 2010. Willow spiling: review of streambank stabilisation projects
essential for N, P and C-retention. Slope lowering and the consequent in the UK. Fr. Rev. 3 (1), 33–47. https://doi.org/10.1608/FRJ-3.1.2.
increase in inundation have major effects on nutrient retention. The BfG & BAW - Federal Institute of Hydrology & Federal Waterways Engineering and
Research Institute, 2008. Untersuchungen zu alternativen technisch-biologischen
highest amount of C-sequestration is found in woody vegetation. To Ufersicherungen an Bundeswasserstraßen, Teil 2: Teststrecke Stolzenau/Weser
meet the current ecological challenges, large-scale river and floodplain [Investigations to alternative technical-biological bank protection measures on
restoration projects should certainly be a priority. However, the avail­ Federal Waterways, Part 2: Study reach Stolzenau/Weser]. Karlsruhe/Koblenz. (BfG-
Report Nr.: 1579).
ability of restoration sites remains a serious challenge. As long as there BMUB - Federal Ministry for the Environment, Nature Conservation and Nuclear Safety,
are no other human demands, the ecological enhancement of riv­ 2015. Naturschutz-Offensive 2020 – Für biologische Vielfalt!. https://www.bmu.de/
erbanks by BRP can be a straightforward step to reactivate floodplain fileadmin/Daten_BMU/Pools/Broschueren/naturschutz-offensive_2020_broschuere_
bf.pdf, Accessed date: 14 November 2018.
ecosystem services. The recurrent need to repair aging infrastructure BMVI- Federal Ministry of Transport and Digital Infrastructure, 2019. Federal Waterways
may be an ideal opportunity for ecological riverbank reconstruction and Shipping Administration (WSV). https://www.bmvi.de/SharedDocs/EN/
(Doyle et al., 2008). We exemplarily show that the potential restoration Articles/WS/federal-waterways-and-shipping-administration.html, Accessed date: 17
February 2019.
of only a small portion of a national river network provides services Brettar, I., Sanchez-Perez, J.M., Trémolieres, M., 2002. Nitrate elimination by deni­
comparable to large scale restoration projects or natural ecosystems. trification in hardwood forest soils of the Upper Rhine floodplain–correlation with
Future restoration attempts such as in the course of the EU- water redox potential and organic matter. Hydrobiologia 469 (1–3), 11–21. https://doi.
org/10.1023/A:1015527611350.
framework directive or the German “Blue Belt Programme” (BMUB,

8
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

Brix, H., Sorrell, B.K., Lorenzen, B., 2001. Are Phragmites-dominated wetlands a net comparative study along German Federal Waterways. Ecol. Eng. 84, 463–475.
source or net sink of greenhouse gases? Aquat. Bot. 69 (2–4), 313–324. https://doi. https://doi.org/10.1016/j.ecoleng.2015.09.019.
org/10.1016/S0304-3770(01)00145-0. Hefting, M.M., Clement, J.C., Bienkowski, P., Dowrick, D., Guenat, C., Butturini, A., ...
Brunotte, E., Dister, E., Günther-Diringer, D., Koenzen, U., Mehl, D., 2009. Flussauen in Verhoeven, J.T., 2005. The role of vegetation and litter in the nitrogen dynamics of
Deutschland–Erfassung und Bewertung des Auenzustandes [Riparian wetlands in riparian buffer zones in Europe. Ecological Engineering 24 (5), 465–482. https://doi.
Germany, inventory and evaluation of the conditions of floodplains]. Naturschutz org/10.1016/j.ecoleng.2005.01.003.
und Biologische Vielfalt 87. Hein, T., Schwarz, U., Habersack, H., Nichersu, I., Preiner, S., Willby, N., Weigelhofer, G.,
Burt, T.P., Matchett, L.S., Goulding, K.W.T., Webster, C.P., Haycock, N.E., 1999. 2016. Current status and restoration options for floodplains along the Danube River.
Denitrification in riparian buffer zones: the role of floodplain hydrology. Hydrol. Sci. Total Environ. 543, 778–790. https://doi.org/10.1016/j.scitotenv.2015.09.073.
Process. 13 (10), 1451–1463. https://doi.org/10.1002/(SICI)1099-1085(199907) Hoffmann, C.C., Kjaergaard, C., Uusi-Kämppä, J., Hansen, H.C.B., Kronvang, B., 2009.
13:10<1451::AID-HYP822>3.0.CO;2-W. Phosphorus retention in riparian buffers: review of their efficiency. J. Environ. Qual.
Canadell, J.G., Raupach, M.R., 2008. Managing forests for climate change mitigation. 38 (5), 1942–1955. https://doi.org/10.2134/jeq2008.0087.
Science 320 (5882), 1456–1457. https://doi.org/10.1126/science.1155458. Hofmann-Schielle, C., Jug, A., Makeschin, F., Rehfuess, K.E., 1999. Short-rotation plan­
Catterall, C.P., Lynch, R., Jansen, A., 2007. Riparian wildlife and habitats. In: Lovett, S., tations of balsam poplars, aspen and willows on former arable land in the Federal
Price, P. (Eds.), Principles for Riparian Lands Management. Land and Water Republic of Germany. I. Site–growth relationships. For. Ecol. Manag. 121 (1–2),
Australia, Canberra, Australia, pp. 141–158. 41–55. https://doi.org/10.1016/S0378-1127(98)00555-6.
Cavaillé, P., Ducasse, L., Breton, V., Dommanget, F., Tabacchi, E., Evette, A., 2015. Jabłońska, E., Wiśniewska, M., Marcinkowski, P., Grygoruk, M., Walton, C.R., Zak, D., ...
Functional and taxonomic plant diversity for riverbank protection works: Kotowski, W., 2020. Catchment-scale analysis Reveals high cost-effectiveness of
Bioengineering techniques close to natural banks and beyond hard engineering. J. Wetland buffer Zones as a remedy to non-point nutrient pollution in North-Eastern
Environ. Manag. 151, 65–75. https://doi.org/10.1016/j.jenvman.2014.09.028. Poland. Water 12 (3), 629. https://doi.org/10.3390/w12030629.
Cierjacks, A., Kleinschmit, B., Babinsky, M., Kleinschroth, F., Markert, A., Menzel, M., ... Jalkanen, A., Mäkipää, R., Ståhl, G., Lehtonen, A., Petersson, H., 2005. Estimation of the
Lang, F., 2010. Carbon stocks of soil and vegetation on Danubian floodplains. Journal biomass stock of trees in Sweden: comparison of biomass equations and age-depen­
of Plant Nutrition and Soil Science 173 (5), 644–653. https://doi.org/10.1002/jpln. dent biomass expansion factors. Ann. For. Sci. 62 (8), 845–851. https://doi.org/10.
200900209. 1051/forest:2005075.
Cohen-Shacham, E., Walters, G., Janzen, C., Maginnis, S., 2016. Nature-Based Solutions to Janssen, P., Cavaillé, P., Bray, F., Evette, A., 2019. Soil bioengineering techniques en­
Address Global Societal Challenges. IUCN, Gland, Switzerland, pp. 97. hance riparian habitat quality and multi-taxonomic diversity in the foothills of the
Collet, D.M., 2004. Willows of interior Alaska. Order 70181 (12-M692). Alps and Jura Mountains. Ecol. Eng. 133, 1–9. https://doi.org/10.1016/j.ecoleng.
Costanza, R., d'Arge, R., De Groot, R., Farber, S., Grasso, M., Hannon, B., ... Raskin, R. G, 2019.04.017.
1997. The value of the world's ecosystem services and natural capital. Nature 387 Kaushal, S.S., Groffman, P.M., Mayer, P.M., Striz, E., Gold, A.J., 2008. Effects of stream
(6630), 253–260. https://doi.org/10.1038/387253a0. restoration on denitrification in an urbanizing watershed. Ecol. Appl. 18 (3),
DeLaune, R.D., Boar, R.R., Lindau, C.W., Kleiss, B.A., 1996. Denitrification in bottomland 789–804. https://doi.org/10.1890/07-1159.1.
hardwood wetland soils of the Cache River. Wetlands 16 (3), 309–320. https://doi. Kayranli, B., Scholz, M., Mustafa, A., Hedmark, Å., 2010. Carbon storage and fluxes
org/10.1007/BF03161322. within freshwater wetlands: a critical review. Wetlands 30 (1), 111–124. https://doi.
Dodds, W.K., 2006. Eutrophication and trophic state in rivers and streams. Limnol. org/10.1007/s13157-009-0003-4.
Oceanogr. 51 (1part2), 671–680. https://doi.org/10.4319/lo.2006.51.1_part_2.0671. Kingsford, R.T., Basset, A., Jackson, L., 2016. Wetlands: Conservation’s poor cousins.
Doyle, M.W., Stanley, E.H., Havlick, D.G., Kaiser, M.J., Steinbach, G., Graf, W.L., ... Aquat. Conserv. Mar. Freshwat. Ecosyst. 26 (5), 892–916. https://doi.org/10.1002/
Riggsbee, J.A., 2008. Aging infrastructure and ecosystem restoration. Science 319 aqc.2709.
(5861), 286–287. https://doi.org/10.1126/science.1149852. Kronvang, B., Hoffmann, C.C., Svendsen, L.M., Windolf, J., Jensen, J.P., Dørge, J., 1999.
Evette, A., Labonne, S., Rey, F., Liebault, F., Jancke, O., Girel, J., 2009. History of Retention of nutrients in river basins. Aquat. Ecol. 33 (1), 29–40. https://doi.org/10.
bioengineering techniques for erosion control in rivers in Western Europe. Environ. 1023/A:1009947907811.
Manag. 43 (6), 972. https://doi.org/10.1007/s00267-009-9275-y. Lautenbach, S., Maes, J., Kattwinkel, M., Seppelt, R., Strauch, M., Scholz, M., Schulz-
Evette, A., Balique, C., Lavaine, C., Rey, F., Prunier, P., 2012. Using ecological and bio­ Zunkel, C., Volk, M., Weinert, J., Dormann, C.F., 2012. Mapping water quality-re­
geographical features to produce a typology of the plant species used in bioengi­ lated ecosystem services: concepts and applications for nitrogen retention and pes­
neering for riverbank protection in Europe. River Res. Appl. 28 (10), 1830–1842. ticide risk reduction. Int. J. Biodivers. Sci. Ecosys. Serv. Manag. 8 (1–2), 35–49.
https://doi.org/10.1002/rra.1560. https://doi.org/10.1080/21513732.2011.631940.
Feld, C.K., Birk, S., Bradley, D.C., Hering, D., Kail, J., Marzin, A., ... Pont, D., 2011. From Lehtonen, A., Mäkipää, R., Heikkinen, J., Sievänen, R., Liski, J., 2004. Biomass expansion
natural to degraded rivers and back again: a test of restoration ecology theory and factors (BEFs) for Scots pine, Norway spruce and birch according to stand age for
practice. Advances in ecological research 44, 119–209. https://doi.org/10.1016/ boreal forests. For. Ecol. Manag. 188 (1), 211–224. https://doi.org/10.1016/j.foreco.
B978-0-12-374794-5.00003-1. 2003.07.008.
Fierke, M.K., Kauffman, J.B., 2005. Structural dynamics of riparian forests along a black Li, M.H., Eddleman, K.E., 2002. Biotechnical engineering as an alternative to traditional
cottonwood successional gradient. For. Ecol. Manag. 215 (1–3), 149–162. engineering methods: a biotechnical streambank stabilization design approach.
Fischenich, J.C., 2003. Effects of Riprap on Riverine and Riparian Ecosystems (No. ERDC/ Landsc. Urban Plan. 60 (4), 225–242. https://doi.org/10.1016/S0169-2046(02)
EL-TR-03-4). Engineer Research and Development Centre Vicksburg MS 00057-9.
Environmental Lab. Li, X., Zhang, L., Zhang, Z., 2006. Soil bioengineering and the ecological restoration of
Fischenich, J.C., Allen, H., 2000. Stream Management (No. ERDC/EL-TR-SR-W-00-1). riverbanks at the Airport Town, Shanghai, China. Ecol. Eng. 26 (3), 304–314. https://
Engineer Research and Development Centre Vicksburg MS Environmental Lab. doi.org/10.1016/j.ecoleng.2005.10.011.
Florsheim, J.L., Mount, J.F., Chin, A., 2008. Bank erosion as a desirable attribute of rivers. Lowrance, R., Altier, L.S., Newbold, J.D., Schnabel, R.R., Groffman, P.M., Denver, J.M., ...
BioScience 58 (6), 519–529. Staver, K.W., 1997. Water quality functions of riparian forest buffers in Chesapeake
Follner, Klaus, Ehlert, Thomas, Neukirchen, Bernd, 2010. The status report on German Bay watersheds. Environmental Management 21 (5), 687–712. https://doi.org/10.
floodplains. 38th IAD Conference 1–5. 1007/s002679900060.
Fortier, J., Truax, B., Gagnon, D., Lambert, F., 2015. Biomass carbon, nitrogen and Matzek, V., Puleston, C., Gunn, J., 2014. Can carbon credits fund riparian forest re­
phosphorus stocks in hybrid poplar buffers, herbaceous buffers and natural woodlots storation? Restor. Ecol. 23 (1), 7–14. https://doi.org/10.1111/rec.12153.
in the riparian zone on agricultural land. J. Environ. Manag. 154, 333–345. https:// McClain, M.E., Boyer, E.W., Dent, C.L., Gergel, S.E., Grimm, N.B., Groffman, P.M., ...
doi.org/10.1016/j.jenvman.2015.02.039. McDowell, W.H., 2003. Biogeochemical hot spots and hot moments at the interface of
Frothingham, K.M., 2008. Evaluation of stability threshold analysis as a cursory method terrestrial and aquatic ecosystems. Ecosystems 6 (4), 301–312. https://doi.org/10.
of screening potential streambank stabilization techniques. Appl. Geogr. 28 (2), 1007/s10021-003-0161-9.
124–133. https://doi.org/10.1016/j.apgeog.2007.07.006. MEA - Millennium Ecosystem Assessment, 2005. Ecosystem and Human Well-Being:
Frouz, J., Dvorščík, P., Vávrová, A., Doušová, O., Kadochová, Š., Matějíček, L., 2015. Wetlands and Waters Synthesis. World Resources Institute, Washington, DC.
Development of canopy cover and woody vegetation biomass on reclaimed and un­ Naiman, R.J., Decamps, H., Pollock, M., 1993. The role of riparian corridors in main­
reclaimed post-mining sites. Ecol. Eng. 84, 233–239. https://doi.org/10.1016/j. taining regional biodiversity. Ecol. Appl. 3 (2), 209–212. https://doi.org/10.2307/
ecoleng.2015.09.027. 1941822.
Gautier, E., Corbonnois, J., Petit, F., Arnaud-Fassetta, G., Brunstein, D., Grivel, S., ... Beck, Naiman, R.J., Decamps, H., McClain, M.E., 2010. Riparia: Ecology, Conservation, and
T., 2009. Multi-disciplinary approach for sediment dynamics study of active flood­ Management of Streamside Communities. Elsevier.
plains. Géomorphologie 15 (1). https://doi.org/10.4000/geomorphologie.7506. Nakamura, K., Tockner, K., Amano, K., 2006. River and wetland restoration: lessons from
Ghestem, M., Sidle, R.C., Stokes, A., 2011. The influence of plant root systems on sub­ Japan. AIBS Bull. 56 (5), 419–429. https://doi.org/10.1641/0006-3568(2006)
surface flow: implications for slope stability. Bioscience 61 (11), 869–879. https:// 056[0419:RAWRLF]2.0.CO;2.
doi.org/10.1525/bio.2011.61.11.6. Natho, S., Venohr, M., Henle, K., Schulz-Zunkel, C., 2013. Modelling nitrogen retention in
Giese, L.A., Aust, W.M., Kolka, R.K., Trettin, C.C., 2003. Biomass and carbon pools of floodplains with different degrees of degradation for three large rivers in Germany. J.
disturbed riparian forests. For. Ecol. Manag. 180 (1–3), 493–508. https://doi.org/10. Environ. Manag. 122, 47–55. https://doi.org/10.1016/j.jenvman.2013.02.049.
1016/S0378-1127(02)00644-8. Newcomer Johnson, T.A., Kaushal, S.S., Mayer, P.M., Smith, R.M., Sivirichi, G.M., 2016.
Gruenewald, H., Brandt, B.K., Schneider, B.U., Bens, O., Kendzia, G., Hüttl, R.F., 2007. Nutrient retention in restored streams and rivers: a global review and synthesis.
Agroforestry systems for the production of woody biomass for energy transformation Water 8 (4), 116. https://doi.org/10.3390/w8040116.
purposes. Ecol. Eng. 29 (4), 319–328. https://doi.org/10.1016/j.ecoleng.2006.09. NIBIS®, 2014. Kartenserver: Bodenübersichtskarte 1:50000 (BUEK50). - Landesamt für
012. Bergbau, Energie und Geologie (LBEG), Hannover.
Harvolk, S., Symmank, L., Sundermeier, A., Otte, A., Donath, T.W., 2015. Human impact Nilsson, C., Svedmark, M., 2002. Basic principles and ecological consequences of chan­
on plant biodiversity in functional floodplains of heavily modified rivers–a ging water regimes: riparian plant communities. Environ. Manag. 30 (4), 468–480.

9
L. Symmank, et al. Ecological Engineering 158 (2020) 106040

https://doi.org/10.1007/s00267-002-2735-2. Nutrient Retention, Carbon Storage, Greenhouse Gas Emissions and Habitat
Nilsson, C., Reidy, C.A., Dynesius, M., Revenga, C., 2005. Fragmentation and flow reg­ Functions]. Naturschutz und biologische Vielfalt, pp. 124.
ulation of the world’s large river systems. Science 308 (5720), 405–408. https://doi. Šimek, M., Cooper, J.E., 2002. The influence of soil pH on denitrification: progress to­
org/10.1126/science.1107887. wards the understanding of this interaction over the last 50 years. Eur. J. Soil Sci. 53
Nordh, N.E., Verwijst, T., 2004. Above-ground biomass assessments and first cutting cycle (3), 345–354. https://doi.org/10.1046/j.1365-2389.2002.00461.x.
production in willow (Salix sp.) coppice—a comparison between destructive and non- Søvik, A.K., Syversen, N., 2008. Retention of particles and nutrients in the root zone of a
destructive methods. Biomass Bioenergy 27 (1), 1–8. https://doi.org/10.1016/j. vegetative buffer zone: effect of vegetation and season. Boreal Environ. Res. 13,
biombioe.2003.10.007. 223–230.
Norris, J.E., Di Iorio, A., Stokes, A., Nicoll, B.C., Achim, A., 2008. Species selection for soil Sutfin, N.A., Wohl, E.E., Dwire, K.A., 2016. Banking carbon: a review of organic carbon
reinforcement and protection. In: Norris, J.E., Stokes, A., Mickovski, S.B., storage and physical factors influencing retention in floodplains and riparian eco­
Cammeraat, E., van Beek, R., Nicoll, B.C., Achim, A. (Eds.), Slope Stability and systems. Earth Surf. Process. Landf. 41 (1), 38–60. https://doi.org/10.1002/esp.
Erosion Control: Ecotechnological Solutions. Springer, Dordrecht. 3857.
Olde Venterink, H., Vermaat, J.E., Pronk, M., Wiegman, F., van der Lee, G.E., van den Sweeney, B.W., Bott, T.L., Jackson, J.K., Kaplan, L.A., Newbold, J.D., Standley, L.J., ...
Hoorn, M.W., ... Verhoeven, J.T., 2006. Importance of sediment deposition and de­ Horwitz, R.J., 2004. Riparian deforestation, stream narrowing, and loss of stream
nitrification for nutrient retention in floodplain wetlands. Applied Vegetation Science ecosystem services. Proceedings of the National Academy of Sciences 101 (39),
9 (2), 163–174. https://doi.org/10.1111/j.1654-109X.2006.tb00665.x. 14132–14137. https://doi.org/10.1073/pnas.0405895101.
Pasche, E., Kräßig, S., Lippert, K., Nasermoaddeli, H., Plöger, W., Rath, S., 2006. Wie viel Symmank, L., Profeta, A., Niens, C., 2020. Valuation of river restoration measures – Do
Physik braucht die Strömungsberechnung in der Ingenieurpraxis? In: residential preferences depend on leisure behavior? Eur. Plan. Stud. https://doi.org/
Wasserbaukolloquium, Strömungssimulation im Wasserbau, Dresdner 10.1080/09654313.2020.1760792. (in press).
Wasserbauliche Mitteilungen, . https://www.researchgate.net/profile/Mhassan_ Tigges, J., Churkina, G., Lakes, T., 2017. Modeling above-ground carbon storage: a re­
Nasermoaddeli/publication/267372681_Wie_viel_Physik_braucht_die_ mote sensing approach to derive individual tree species information in urban settings.
Stromungsberechnung_in_der_Ingenieurpraxis/links/5627784408ae2b313c54e63f. Urban Ecosyst. 20 (1), 97–111. https://doi.org/10.1007/s11252-016-0585-6.
pdf. Tisserant, M., Janssen, P., Evette, A., González, E., Cavaillé, P., Poulin, M., 2020. Diversity
Penman, J., Gytarsky, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., ... Wagner, F., and succession of riparian plant communities along riverbanks bioengineered for
2003. Good practice guidance for land use, land-use change and forestry. Institute for erosion control: a case study in the foothills of the Alps and the Jura Mountains. Ecol.
Global Environmental Strategies, Hayama, Kanagawa, Japan Available at: http:// Eng. 152, 105880.
www.80.24.165.149/webproduccion/PDFs/03CAP05.PDF. Tockner, K., Stanford, J.A., 2002. Riverine flood plains: present state and future trends.
Pinay, G., Clément, J.C., Naiman, R.J., 2002. Basic principles and ecological con­ Environ. Conserv. 29 (3), 308–330. https://doi.org/10.1017/S037689290200022X.
sequences of changing water regimes on nitrogen cycling in fluvial systems. Environ. Trepel, M., Palmeri, L., 2002. Quantifying nitrogen retention in surface flow wetlands for
Manag. 30 (4), 481–491. https://doi.org/10.1007/s00267-002-2736-1. environmental planning at the landscape-scale. Ecol. Eng. 19 (2), 127–140. https://
Pinto, A., Fernandes, L.S., Maia, R., 2016. Monitoring methodology of interventions for doi.org/10.1016/S0925-8574(02)00038-1.
riverbanks stabilization: Assessment of technical solutions performance. Water Tufekcioglu, A., Raich, J.W., Isenhart, T.M., Schultz, R.C., 2003. Biomass, carbon and
Resour. Manag. 30 (14), 5281–5298. https://doi.org/10.1007/s11269-016-1486-4. nitrogen dynamics of multi-species riparian buffers within an agricultural watershed
Rheinhardt, R.D., Brinson, M.M., Meyer, G.F., Miller, K.H., 2012. Carbon storage of in Iowa, USA. Agrofor. Syst. 57 (3), 187–198. https://doi.org/10.1023/
headwater riparian zones in an agricultural landscape. Carbon Balance Manag. 7 (1), A:1024898615284.
4. https://doi.org/10.1186/1750-0680-7-4. UNEP-WCMC - World Conservation Monitoring Center, 2000. European Forests and
Rieger, I., Lang, F., Kowarik, I., Cierjacks, A., 2014. The interplay of sedimentation and Protected Areas: Gap Analysis. United Nations Environmental Program - World
carbon accretion in riparian forests. Geomorphology 214, 157–167. https://doi.org/ Conservation Monitoring Centre, Cambridge, UK.
10.1016/j.geomorph.2014.01.023. Verhoeven, J.T., Arheimer, B., Yin, C., Hefting, M.M., 2006. Regional and global concerns
Ritz, S., Dähnke, K., Fischer, H., 2018. Open-channel measurement of denitrification in a over wetlands and water quality. Trends Ecol. Evol. 21 (2), 96–103. https://doi.org/
large lowland river. Aquat. Sci. 80 (1), 11. https://doi.org/10.1007/s00027-017- 10.1016/j.tree.2005.11.015.
0560-1. Vidon, P., 2010. Riparian zone management and environmental quality: a multi‐­
Robinson, C.T., Tockner, K., Ward, J.V., 2002. The fauna of dynamic riverine landscapes. contaminan challenge. Hydrological Processes. Int. J. 24 (11), 1532–1535.
Freshw. Biol. 47 (4), 661–677. https://doi.org/10.1046/j.1365-2427.2002.00921.x. von Haaren, C., SCHRÖTER-SCHLAACK, C., ALBERT, C., HANSJÜRGENS, B., & KRÄTZIG,
Rytter, R.M., 2012. The potential of willow and poplar plantations as carbon sinks in S. A. (2016). Ökosystemleistungen in ländlichen Räumen. Grundlage für mens­
Sweden. Biomass Bioenergy 36, 86–95. https://doi.org/10.1016/j.biombioe.2011. chliches Wohlergehen und nachhaltige wirtschaftliche Entwicklung. Hannover,
10.012. Leipzig: Naturkapital Deutschland-TEEB DE.
Sala, O.E., Chapin, F.S., Armesto, J.J., Berlow, E., Bloomfield, J., Dirzo, R., ... Leemans, R., Walton, C.R., Zak, D., Audet, J., Petersen, R.J., Lange, J., Oehmke, C., ... Kotowski, W.,
2000. Global biodiversity scenarios for the year 2100. Science 287 (5459), 2020. Wetland buffer zones for nitrogen and phosphorus retention: Impacts of soil
1770–1774. https://doi.org/10.1126/science.287.5459.1770. type, hydrology and vegetation. Science of The Total Environment 138709. https://
Schmitt, K., Schäffer, M., Koop, J., Symmank, L., 2018. Riverbank stabilisation by doi.org/10.1016/j.scitotenv.2020.138709.
bioengineering: potentials for ecological diversity. J. Appl. Water Eng. Res. 1–12 Waters, E.R., Morse, J.L., Bettez, N.D., Groffman, P.M., 2014. Differential carbon and
(1080/23249676.2018.1466735). nitrogen controls of denitrification in riparian zones and streams along an urban to
Schneider, S., Nestmann, F., Lehmann, B., 2006. Interaction of vegetation, current and exurban gradient. J. Environ. Qual. 43 (3), 955–963. https://doi.org/10.2134/
sedimantation. In: River Flow 2006, Portugal, pp. 647–656. jeq2013.12.0504.
Scholz, M., Kasperidus, H.D., Ilg, C., Henle, K., Kasperidus, H.D., Born, W., Henle, K., Wells, Gary W., "Biotechnical Streambank Protection: The use of plants to stabilize
2012. Habitatfunktion (Habitat function). In: Scholz, M., Mehl, D., Schulz-Zunkel, C. streambanks" (2002). Agroforestry Notes (USDANAC). 23.
(Eds.), Ökosystemfunktionen von Flussauen e Analyse und Bewertung von Wollny, J.T., Otte, A., Harvolk-Schöning, S., 2019. Dominance of competitors in riparian
Hochwasserretention, Nährstoffrückhalt, Kohlenstoffvorrat, Treibhausgasemissionen plant species composition along constructed banks of the German rivers Main and
und Habitatfunktion. [Ecosystem services of riparian Floodplains-Analysis and Danube. Ecol. Eng. 127, 324–337. https://doi.org/10.1016/j.ecoleng.2018.11.013.
Evaluation of Flood Retention, Nutrient Retention, Carbon Storage, Greenhouse Gas Wolter, C., 2001. Conservation of fish species diversity in navigable waterways. Landsc.
Emissions and Habitat Functions]. Naturschutz und biologische Vielfalt, pp. 124. Urban Plan. 53 (1–4), 135–144. https://doi.org/10.1016/S0169-2046(00)00147-X.
Schultz, R.C., Isenhart, T.M., Colletti, J.P., 2005. Riparian buffer systems in crop and Wu, H.L., Feng, Z.Y., 2006. Ecological engineering methods for soil and water con­
rangelands. In: Natural Resource Ecology and Management Conference Papers, servation in Taiwan. Ecol. Eng. 28 (4), 333–344. https://doi.org/10.1016/j.ecoleng.
Posters and Presentations. 23. http://lib.dr.iastate.edu/nrem_conf/23. 2006.09.005.
Schulz-Zunkel, C., Scholz, M., Kasperidus, H.D., Krüger, F., Natho, S., Venohr, M., 2012. Zerbe, S., Steffenhagen, P., Parakenings, K., Timmermann, T., Frick, A., Gelbrecht, J., Zak,
Nährstoffrückhalt (Nutrient retention). In: Scholz, M., Mehl, D., Schulz-Zunkel, C., D., 2013. Ecosystem service restoration after 10 years of rewetting peatlands in NE
Kasperidus, H.D., Born, W., Henle, K. (Eds.), Ökosystemfunktionen von Flussauen e Germany. Environ. Manag. 51 (6), 1194–1209. https://doi.org/10.1007/s00267-
Analyse und Bewertung von Hochwasserretention, Nährstoffrückhalt, 013-0048-2.
Kohlenstoffvorrat, Treibhausgasemissionen und Habitatfunktion. [Ecosystem Zianis, D., Muukkonen, P., Mäkipää, R., Mencuccini, M., 2005. Biomass and stem volume
Services of Riparian Floodplains-Analysis and Evaluation of Flood Retention, equations for tree species in Europe. Silva Fenn. Monogr. 4, 1–63.

10