HESSHydrology and Earth System SciencesHESSHydrol. Earth Syst. Sci.1607-7938Copernicus PublicationsGöttingen, Germany10.5194/hess-21-5763-2017Importance of considering riparian vegetation requirements for the long-term
efficiency of environmental flows in aquatic microhabitatsRivaesRuiruirivaes@isa.ulisboa.pthttps://orcid.org/0000-0001-7910-4387BoavidaIsabelSantosJosé M.https://orcid.org/0000-0003-1232-2560PinheiroAntónio N.https://orcid.org/0000-0002-1423-5391FerreiraTeresahttps://orcid.org/0000-0002-3900-1460Forest Research Centre, Instituto Superior de Agronomia,
Universidade de Lisboa, Tapada da Ajuda 1349-017 Lisbon,
PortugalCERIS, Civil Engineering Research Innovation and
Sustainability Centre, Instituto Superior Técnico,
Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon, PortugalRui Rivaes (ruirivaes@isa.ulisboa.pt)22November20172111576357802February201721March20177August201713October2017This work is licensed under the Creative Commons Attribution 3.0 Unported License. To view a copy of this licence, visit https://creativecommons.org/licenses/by/3.0/This article is available from https://hess.copernicus.org/articles/21/5763/2017/hess-21-5763-2017.htmlThe full text article is available as a PDF file from https://hess.copernicus.org/articles/21/5763/2017/hess-21-5763-2017.pdf
Environmental flows remain biased toward the traditional
biological group of fish species. Consequently, these flows ignore the
inter-annual flow variability that rules species with longer lifecycles and
therefore disregard the long-term perspective of the riverine ecosystem. We
analyzed the importance of considering riparian requirements for the
long-term efficiency of environmental flows. For that analysis, we modeled
the riparian vegetation development for a decade facing different
environmental flows in two case studies. Next, we assessed the corresponding
fish habitat availability of three common fish species in each of the
resulting riparian landscape scenarios. Modeling results demonstrated that
the environmental flows disregarding riparian vegetation requirements
promoted riparian degradation, particularly vegetation encroachment. Such
circumstance altered the hydraulic characteristics of the river channel where
flow depths and velocities underwent local changes of up to 10 cm and
40 cm s-1, respectively. Accordingly, after a decade of this flow
regime, the available habitat area for the considered fish species
experienced modifications of up to 110 % when compared to the natural
habitat. In turn, environmental flows regarding riparian vegetation
requirements were able to maintain riparian vegetation near natural
standards, thereby preserving the hydraulic characteristics of the river
channel and sustaining the fish habitat close to the natural condition. As a
result, fish habitat availability never changed more than 17 % from the
natural habitat.
Introduction
Freshwater ecosystems provide vital services for human existence but are on
top of the world's most threatened ecosystems (Dudgeon et al., 2006; Revenga
et al., 2000), primarily due to river damming (Allan and Castillo, 2007). The
ability to provide sufficient water to ensure the functioning of freshwater
ecosystems is an important concern as its capacity to provide goods and
services is sustained by water-dependent ecological processes (Acreman,
2001). The relevance of this subject compelled the scientific community to
appeal to all governments and water-related institutions across the globe to
engage in environmental flow restoration and maintenance in every river
(Brisbane Declaration, 2007). Actually, this issue is a global
research topic, as all dams, weirs, and levees
change the magnitudes of peak flood flows of rivers to a certain extent
(e.g., FitzHugh and Vogel, 2010; Maheshwari et al., 1995; Miller et al.,
2013; Nilsson and Berggren, 2000; Uddin et al., 2014a, b). As a result of
this, there are still opportunities for the implementation of environmental
flow restoration at hundreds of thousands of these structures worldwide
(Richter and Thomas, 2007).
Environmental flows can be defined as “the quantity, timing and quality of
water flows required to sustain freshwater and estuarine ecosystems, and the
human livelihoods and wellbeing that depend upon these ecosystems” (Brisbane
Declaration, 2007) and play an essential role in the conservation of
freshwater ecosystems (Arthington et al., 2006; Hughes and Rood, 2003). It is
now agreed that environmental flows must ideally be based on the ecological
requirements of different biological communities (e.g., Acreman et al., 2009,
2014; Acreman and Ferguson, 2010; Arthington et al., 2010; Arthington, 2012;
Arthington and Zalucki, 1998; Davis and Hirji, 2003; Dyson et al., 2003; Poff
et al., 1997) and should present a dynamic and variable hydrological regime
to maintain the native biodiversity and the ecological processes that
represent every river (Bunn and Arthington, 2002; Lytle and Poff, 2004;
Postel and Richter, 2003). In this sense, holistic methodologies meant to
address river systems as a whole (Arthington et al., 1992; King and Tharme,
1994; King and Louw, 1998) are clearly being increasingly applied out of
Australia and South Africa (Hirji and Davis, 2009), the origin countries of
this holistic concept. However, the most commonly applied methods throughout
the world are still hydrologically based methods (Dyson et al., 2003;
Linnansaari et al., 2012; Tharme, 2003). Conversely, environmental flows
ascertained through habitat simulation methods still persist generally based
on the requirements of a single biological group, mostly fish (Acreman et
al., 2009; Arthington, 2012; Tharme, 2003), and require an input from less
typically monitored taxa (Gillespie et al., 2014). Accordingly, these
approaches still disregard the inter-annual flow variability that rules
species with longer lifecycles, like riparian vegetation, therefore lacking
the long-term perspective of the riverine ecosystem (Stromberg et al., 2010).
The feedbacks of these shortcomings on the riparian and aquatic communities
were seldom estimated before and so, the efficiency of such approaches along
with its long-term after-effects remains practically unknown.
Riparian vegetation is a suitable environmental change indicator (Benjankar
et al., 2012; Nilsson and Berggren, 2000) that responds directly to a flow
regime in an inter-annual time frame (Capon and Dowe, 2007; Naiman et al.,
2005; Poff et al., 1997) and has a clear significance in the habitat
improvement of aquatic systems (e.g., Broadmeadow and Nisbet, 2004; Chase et
al., 2016; Dosskey et al., 2010; Gregory et al., 1991; Pusey and Arthington,
2003; Rood et al., 2015; Ryan et al., 2013; Salemi et al., 2012; Statzner,
2012; Tabacchi et al., 2000; Van Looy et al., 2013; Wootton, 2012). In fact,
riparian vegetation and aquatic species interact biologically, physically,
and chemically (Gregory et al., 1991). Riparian vegetation is capable of
influencing aquatic species in several ways. It affects food webs by
providing an important input of nutrients that are a major food source for
invertebrates, which are in turn eaten by fishes (Wootton, 2012). It
influences hydrological processes (Salemi et al., 2012; Tabacchi et al.,
2000) and protects aquatic habitats by means of river bank stability (Rood et
al., 2015) and providence of large woody debris (Fetherston et al., 1995). It
provides thermal regulation of rivers by overshadowing (Ryan et al., 2013)
and protects water quality both by trapping sediments and contaminants (Chase
et al., 2016) as by chemical uptake and cycling (Dosskey et al., 2010). On
the other hand, aquatic species also appear to be able to influence riparian
zones, although at a much smaller magnitude, acting as ecosystem engineers
(Statzner, 2012). For instance, fishes can dig in sand and gravel for food or
reproductive purposes and therefore influence sediment surface
characteristics and critical shear stress (e.g., Hassan et al., 2008;
Statzner et al., 2003).
Accordingly, riparian restoration is an indispensable implementation measure
to recover the natural river processes and is the most promising restoration
action in many degraded rivers (Palmer et al., 2014). Hence, incorporating
riparian vegetation requirements (the need for specific flows to preserve
the naturalness of recruitment and meta-stability facing fluvial processes)
into environmental flows could be an important contribution to fill in these
gaps.
We have already noticed how environmental flow regimes disregarding riparian
vegetation requirements allow for the degradation of riparian woodlands in
the subsequent years following such river regulation (e.g., Rivaes et al.,
2015). However, we are not aware of studies assessing the return effect of
this degradation again on the efficiency of those environmental flow regimes.
The purpose of this study is to evaluate the effect of disregarding riparian
vegetation requirements in the efficiency of environmental flow regimes
regarding fish habitat availability in the long-term perspective of the
fluvial ecosystem. We used an approach from an ecohydraulic point of view to
evaluate the effects of riparian landscape degradation on fish species. By
riparian landscape we mean the specific spatial patterns of riparian
vegetation that result from ecological, geomorphological, and hydrological
processes and are depicted by the existing patch mosaic with different
vegetation types and succession phases. We were particularly interested in
answering the following questions: (i) are environmental flows exclusively
addressing fish requirements capable of preserving the habitat availability
of these aquatic species in the long term? (ii) If not, to what extent can
the disregard for riparian vegetation requirements derail the goals of
environmental flows addressing only aquatic species as a result of the
riparian landscape degradation? (iii) Are environmental flows regarding
riparian requirements able to maintain the habitat availability of fish
species?
To approach these questions, we first modeled the structural response of
riparian vegetation (please see Naiman et al., 2005, and NRC, 2002, for a
better understanding about riparian vegetation structure) facing a decade of
different environmental flows in two different case studies. Next, we
performed an assessment of habitat availability for fish species in each of
the resulting riparian landscape scenarios. We are not aware of such a
modeling approach ever being used in the appraisal of the long-term
efficiency of environmental flow regimes, which can provide an extremely
valuable insight into the expected long-term effects of environmental flows
in river ecosystems in advance.
Location and characterization of the study sites OCBA and OCPR.
MethodsStudy sites
The two study sites were selected in the Ocreza River, eastern Portugal
(Fig. 1). This is a medium-sized stream that runs on schistose rocks for
94 km and drains a 1429 km2 watershed with a mean annual flow of
16.5 m3 s-1. The flow regime is typically Mediterranean (Gasith
and Resh, 1999), with a low flow period interrupted by flash floods in winter
(the median of mean daily discharges in the winter months is
8.8 m3 s-1 and maximum annual discharges with return periods of
2, 5, 10, and 100 years are, respectively, 323, 549, 718, and
1314 m3 s-1) and a very low flow, even null at times, during
summer (the first quartile and the median of mean daily discharges in the
summer months are, respectively, 0 and 0.1 m3 s-1). Two study
sites were considered (OCBA and OCPR) to provide a broader analysis of the
aquatic habitat modifications in different hydrogeomorphological contexts.
The OCBA study site (39∘44′07.05′′ N,
7∘44′16.51′′ W) is located 30 km upstream from the river mouth
and OCPR (39∘43′16.88′′ N, 7∘46′01.05′′ W) is
approximately 5 km downstream of OCBA. Despite the relatively small distance
between them, several characteristics differentiate the two study sites.
While in OCBA, the river flows freely on a boulder substrate and is confined
to steep valley hillsides, in OCPR, the river flows on a coarser boulder
substrate with sparse bedrock presence and is located in a relatively wider
valley section. OCBA and OCPR also differ in watershed areas, representing 54
and 72 % of the entire river basin, respectively. This feature further
differentiates the two case studies, as the intermediate watershed of OCPR
collects water from a much rainier zone, thereby conferring an increased flow
regime in this study site. The surveyed areas in the OCBA and OCPR study
sites encompass a river length of approximately 500 and 300 m, respectively,
laterally limited by the 100-year flooded zone, thus totaling approximately 4
and 3 ha for the OCBA and OCPR study sites, respectively. In both cases, the
fish community is characterized by native cyprinid species, mainly
Luciobarbus bocagei (Iberian barbel, hereafter barbel),
Pseudochondrostoma polylepis (Iberian straight-mouth nase, hereafter
nase), and Squalius alburnoides (calandino), whereas the local
riparian vegetation is composed mostly of willows (Salix salviifolia
Brot. and Salix atrocinerea Brot.) and ashes (Fraxinus angustifolia Vahl).
Data collectionHydraulic data
The riverbed topography was surveyed in 2013 using a combination of a Nikon
DTM330 total station and a Global Positioning System (GPS) (Ashtech, model
Pro Mark2). Altogether, 7707 points were surveyed at OCBA and 25 132 at
OCPR. Trees, boulders, and large objects emerging from the water were defined
by marking the object intersection with the riverbed and by surveying the
points necessary to approximately define its shape.
Hydraulic data, i.e., water velocities and depths, were measured as a series
of points along several cross sections in the study sites. Depths were
measured with a ruler and water velocities with a flow probe (model 002,
Valeport) positioned at 60 % of the local depth below the surface (Bovee
and Milhous, 1978). Additionally, the substrate composition was visually
assessed and mapped to determine posteriorly the effective roughness heights
of the riverbed. These data were used to calculate river discharge in each
study site and to calibrate the model. Additional information about hydraulic
data and channel bed characteristics is provided as the Supplement (Sect. S1
– Tables S1, S2, S3, and S4).
Riparian vegetation data
The riparian vegetation was assessed in 2013 to support the calibration and
validation of the riparian vegetation model. This task consisted in recording
the location and shape of all homogeneous vegetation patches with a sub-meter
precision handheld GPS (Ashtech, Mobile Mapper 100), while
dendrochronological methods were used to determine the approximate age of the
patches. Two or three of the largest individuals in each patch were cored
with a standard 5 mm increment borer, taking two perpendicular cores at
breast height in adult trees (Mäkinen and Vanninen, 1999). For
individuals with a diameter smaller than 5 cm at breast height, discs were
obtained for age calculation purposes, and on multistemmed trees, the
cores/discs were taken from the largest stem. The patches were later
classified by succession phase according to its corresponding development
stage. Patch georeferencing, patch aging, and succession phase classification
followed the methodology used by Rivaes et al. (2013).
Five succession phases were identified in the study sites: Initial phase
(IP), Pioneer phase (PP), Early Successional Woodland phase (ES), Established
Forest phase (EF), and Mature Forest phase (MF). Initial phase was attributed
to all patches dominated by gravel bars, sometimes covered by herbaceous
vegetation but without woody arboreal species. The patches dominated by the
recruitment of woody arboreal species were considered to be the Pioneer
phase. The Early Successional Woodland phase classification was attributed to
all patches with a high standing biomass and well-established individuals,
dominated by pioneer watertable-dependent species, such as willows and alders
(Alnus glutinosa). Older patches dominated by macrophanerophytes,
such as ash trees, were considered to be in the Established Forest phase. The
Mature Forest phase was considered at patches where terrestrial vegetation
was also present, determining the transition phase to the upland vegetation
communities. Further information on the characterization of succession phases
is provided as the Supplement (Sect. S2 – Table S6 and Figs. S1 and S2).
Fish data
Fish populations were sampled during 2012 and 2013 at undisturbed or
minimally disturbed sites in the Ocreza basin, an essential requisite when
studying habitat preferences of stream fishes in order to reflect their
optimal habitat (Gorman and Karr, 1978). Sampling occurred in autumn
(November 2012), spring (May 2013) and early summer (June 2013) when there is
full connectivity among instream habitats. Overall, four native species
(cyprinids) were found – barbel, nase, calandino and the Southern Iberian
chub (Squalius pyrenaicus). The latter was however excluded from the
present study, as an insufficient number of individuals were collected to
draw unbiased conclusions. Non-native fish (the gudgeon Gobio lozanoi) occurred in the study area, but in very low density. Field
procedures followed those by Boavida et al. (2011, 2015). Fish sampling was
performed during daylight using pulsed DC electrofishing (SAREL model
WFC7-HV; Electracatch International, Wolverhampton, UK), with low voltage
(250 V) and a 30 cm diameter anode to reduce the effect of positive
galvanotaxis. A 200 m long reach at each site was surveyed by wading
upstream in a zigzag pattern to ensure full coverage of available habitats.
To avoid displacements of individuals from their original positions, a
modified point electrofishing procedure was employed (Copp, 1989). Sampling
points were approached discreetly, and the activated anode was swiftly
immersed in the water for five seconds. Upon sighting a fish or a shoal of
fishes, a numbered location marker was anchored to the streambed for
subsequent microhabitat use measurements. Fish were immediately collected by
means of a separate dip net held by another operator, quickly measured for
total length (TL), and then placed in buckets with portable ELITE aerators to
avoid continuous shocking and repeated counting, before being returned alive
to the river. Ensuing fish sampling, microhabitat measurements of flow depth
(cm), mean water velocity (cm s-1) and dominant substrate composition
were taken in 0.8 by 0.8 m quadrats at the location where each fish was
captured. Microhabitat availability measurements were made using the same
variables by quantifying randomly selected points along 15–25 m equidistant
transects perpendicular to the flow at each sampling site. To develop habitat
suitability curves (HSCs) for target fish size classes, microhabitat
variables (flow depth, water velocity, dominant substrate and cover) were
divided into classes, and histograms of frequencies of use and availability
were constructed (Boavida et al., 2011). A summary of collected fish data, as
well as data analysis to determine habitat use, availability, and preference
of fish species regarding the considered variables, is provided as the
Supplement (Sect. S2 – Table S7 and Figs. S3 to S12).
Flow regime definition
Three flow regimes were considered for the modeling of riparian vegetation:
(i) the natural flow regime (hereafter named natural flow regime), (ii) an
environmental flow regime considering only fish requirements (hereafter named
the Eflow regime), and (iii) an environmental flow regime considering both
fish and riparian requirements (hereafter named the Eflow&Flush regime).
The natural flow regime data was obtained from the Portuguese Water Resources
National Information System (SNIRH, 2010). The environmental flow regimes
used in this study are an adaptation from the environmental flow regime
created by Ferreira et al. (2014) for the location of the study sites
(Fig. 2). These authors determined an environmental flow regime presented in
a multiannual fashion considering a decadal time frame and accounting for two
different flow regime components: a monthly flow regime addressing fish
requirements and a multiannual flow regime composed by floods with different
recurrence intervals addressing riparian vegetation requirements. The first
component, i.e., the flow regime addressing fish requirements (Eflow), was
determined according to the instream flow incremental methodology (Bovee,
1982) and was built on a monthly basis to embody the intra-annual variability
ruling the main lifecycle events of this biological group (Encina et al.,
2006; Gasith and Resh, 1999). These mean monthly discharges addressing fish
requirements that compose the Eflow aimed for the following
goals: (i) maximize the habitat of the target species while attributing the
same weight for each species; (ii) privilege the spawning months (spring;
Santos et al., 2005) and promote the younger life stages during
summer; (iii) maintain the characteristic intra-annual variability of the
river flow; and (iv) preserve the natural regime whenever the environmental
flows suggest higher discharges. The second component of the environmental
flow regime (floods with a certain recurrence interval) proposed by Ferreira
et al. (2014) was determined according to Rivaes et al. (2015) and intends to
characterize the inter-annual flow variability to which the arrangement of
riparian vegetation communities respond (Hughes, 1997). The flushing flows
addressing riparian requirements in the Eflow&Flush regime were defined
based on the need of riparian communities for the minimum necessary flushing
flow regime to maintain the viability and sustainability of riparian
vegetation, particularly, avoiding vegetation encroachment and conserving the
ecological succession equilibrium of the riparian ecosystem (Rivaes et al.,
2015). Therefore, the environmental flow regimes used in this study are
considered an adaptation from Ferreira et al. (2014) as we used just the
fish-addressing component (only mean monthly discharges) as the standard
procedure of an environmental flow regime considering only fish requirements
(Eflow) and both components (mean monthly discharges and flushing flows) for
the environmental flow regime addressing fish and riparian requirements
(Eflow&Flush).
Environmental flow regime addressing fish (black line, left axis)
and riparian (grey bars, right axis) requirements considered for the habitat
modeling in the OCBA study site. Fish requirements are addressed by a
constant monthly discharge and riparian requirements by a flushing flow in
the years in which they are planned (the
duration of the flushing flow is similar to a natural flood with an equal
recurrence interval). The hydrograph for the Eflow&Flush flow regime is
similar in the OCPR study site.
Riparian vegetation modeling
The riparian vegetation modeling was performed using the
CASiMiR-vegetation model (Benjankar et al., 2009). This tool
simulates the succession dynamics of riparian vegetation, based on the
existing relationships of the ecological relevant hydrological elements (Poff
et al., 1997) and the vegetation metrics that reflect riparian communities to
such hydrological alterations (Merritt et al., 2010). The strengths of this
model are the capacity to incorporate the past patch dynamics into every
model run, the ability to work at a response guild level by using succession
phases as modeling units, and the ability to provide the outputs in a
spatially explicit way. In turn, the main disadvantages of this model can be
attributed to the non-existence of a plant competition module or the lack of
an incorporated hydrodynamic model.
Maximum annual discharges (m3 s-1) considered in the
CASiMiR-vegetation model for each study site.
The rationale of this model is based on the fact that riparian communities
respond to the hydrological and habitat variations on a timescale between the
year and the decade (Frissell et al., 1986; Thorp et al., 2008), being that
the flood pulse is the predominant factor in these population dynamics (Thoms
and Parsons, 2002). For these reasons, the hydrological regime is inputted
into the model in terms of maximum annual discharges as these discharges are
considered to be the annual threshold for riparian morphodynamic disturbance
that determine the succession or retrogression of vegetation. Notwithstanding
this, the model also predicts the annual riparian adjustments according to
its vital rates in relation to groundwater depth, as well as the annual
recruitment areas, based on the annual minimum mean daily discharges. The
groundwater depth corresponding to the mean annual discharge of the river is
also a model input used as a reference for the general habitat conditions
that determine the expected riparian landscape according to the calibrated
thresholds of the riparian succession phases. Thus, the magnitude and
duration of extreme low flows are accounted for by the CASiMiR-vegetation
model. A complete detailing of model rationale and parameterization can be
found in Politti and Egger (2011) and Benjankar et al. (2011). Model
calibration was carried out in accordance with the methodology described in
previous studies (García-Arias et al., 2013; Rivaes et al., 2013). In
particular, calibration was performed by running the CASiMiR-vegetation model
for a decade to simulate the effect of the local historic flow regime on
riparian vegetation. The result of the model was then compared with an
observed vegetation map that was surveyed in the same year as the one
corresponding to the result of the model. This is an iterative process of
trial and error where the parameter of the shear stress resistance threshold
of each succession phase is tuned to obtain the best calibration outcome (see
Wainwright and Mulligan, 2004, for a better understanding). All the other
parameters, namely, patch age and height above water table ranges, were
determined based on the data collected in the field. This information is
provided as the Supplement (Sect. S1 – Table S5). During calibration, the riparian
vegetation model achieved an agreement evaluation of 0.61 by the quadratic
weighted kappa (Cohen, 1960), which is considered to be in good agreement
with the observed riparian landscape (Altman, 1991; Viera and Garrett, 2005).
This agreement evaluation can be understood as a classification 61 %
better than what would be expected by a random assignment of classes. The
riparian vegetation model was further validated in this specific watershed
(Ferreira et al., 2014), with even better results (quadratic weighted kappa
of 0.68). After calibration and validation (calibrated parameters provided as
the Supplement; Sect. S1 – Table S5), the riparian vegetation was modeled
for periods of 10 years according to the corresponding flow regimes
(Table 1). Such a modeling period was considered to be long enough to avoid
the influence of the initial vegetation conditions, while river morphological
changes still do not assume importance in vegetation development (Politti et
al., 2014). Furthermore, during modeling, riverbed topography was considered
fixed for several reasons: the study sites are located in a fairly steep
valley in which the river is not allowed to meander considerably during such
a short timescale; the typical substrate of both study sites is
armored and very coarse (boulders,
large boulders, and bedrock); in these conditions the small monthly
discharges intended to maintain aquatic fauna requirements are not able to
create water depths and flow velocities capable of moving or eroding
particles of the size of those found as substrate in the considered study
sites (for a better understanding, please see Alexander and Cooker, 2016;
Clarke and Hansen, 1996; Hjulström, 1939); no significant differences
were found during the substrate analysis of the different succession phases;
prior knowledge of the authors shows that the considered floods do not bring
noteworthy changes to river geomorphology during this period (Rivaes et al.,
2015); the model calibration and validation results exhibited a good
agreement with the observed riparian landscape while using the same
methodology; and by using a fixed topography it is possible to analyze the
exclusive effect of riparian landscape degradation on the river hydraulics.
The resulting riparian vegetation maps were then used as the respective
riparian landscapes (hereafter named the natural, Eflow, and Eflow&Flush
landscapes) in the hydrodynamic modeling of the fish habitat in each study
site.
Hydrodynamic modeling of fish habitats
The hydrodynamic modeling was performed using a calibrated version of the
River2D model (Steffler et al., 2002). This is a finite element model widely
used in fluvial modeling studies for the assessment of habitat availability
(Boavida et al., 2011; Jalón and Gortázar, 2007) that brings together
a 2-D hydrodynamic model and a habitat model to simulate the flow conditions
of the river stretch and estimate its potential habitat value according to
the fish habitat preferences. The strengths of this model are the fact of
being public domain software and being technically robust throughout a wide
range of modeling circumstances. On the other hand, some limitations of this
model are the non-incorporation of a morphodynamic module or the ability to
embody fuzzy logic rules during the computation of species habitat
availability.
The calibration procedure followed the methodology proposed by Boavida et
al. (2013, 2015). Calibration was performed by iteratively adjusting the bed
channel roughness to attain a good agreement of the simulated versus surveyed
water surface elevations and velocity profiles in the surveyed cross
sections. Boundary conditions were set according to the water surface
elevations measured at the upstream and downstream cross sections. Calibrated
parameters are provided in the Supplement (Sect. S1 – Tables S1, S2, S3,
and S4).
The hydrodynamic modeling comprised the Eflow discharge ranges in the study
sites (0–2 and 0–5.5 m3 s-1 for OCBA and OCPR, respectively)
and was accomplished for each riparian landscape scenario. The different
riparian landscapes were represented in the hydrodynamic model by changing
the channel roughness according to the spatial extent of the riparian
succession phases; i.e., the channel roughnesses inputted to the model are
the riparian landscape maps converted into channel roughness maps. Roughness
is a critical feature influencing the physical variables of flow hydraulics
(Chow, 1959; Curran and Hession, 2013), whose distinct combinations typify
diverse functional habitats, which are selected by fish according to its
preference. The roughness classification of riparian vegetation succession
phases was determined based on the roughness measurement literature on
similar vegetation types (Chow, 1959; Wu and Mao, 2007) and expert judgment
during model calibration.
After modeling the Eflow discharges in each of the riparian landscape
scenarios of the two study sites, the hydraulic characteristics of each
riparian landscape (roughness, flow depth and velocity) were compared using
a t-test (confidence level of 99 %) in R environment (R Development Core
Team, 2011) in order to determine the existence of mean significant
differences between riparian landscapes. Habitat simulation was achieved by
the combination of the hydraulic modeling (flow depth and velocity) with
preference curves information for the considered target species. The
riverbed characteristics of substrate and cover were kept unchanged during
the hydrodynamic modeling. Changing the substrate according to the
modifications in succession phase disposal seemed to be an incorrect
practice in this case because during data treatment, no significant
differences were detected in riverbed substrate between succession phases.
Cover modification was also disregarded because the CASiMiR-vegetation model
only reproduces the riparian area, not the aquatic zone (note that this
aquatic zone is a definition sensu CASiMiR-vegetation model, designating the area of the
river channel that is permanently submerged throughout the hydrologic year
and where riparian vegetation is unable to establish and develop. It
corresponds to only a fraction of the wetted area by river flow during the
discharges considered in the subsequent hydrodynamic modeling.) and
therefore, this feature cannot be correctly modeled by the riparian
vegetation model. Notwithstanding, the most important variables determining
fish habitat availability influenced by riparian vegetation degradation were
considered, namely, depth, velocity and substrate (Parasiewicz, 2007).
The Habitat Suitability Index (HSI) was determined for each species and life
stage regarding the product of the velocity (Velocity Suitability Index –
VSI), depth (Depth Suitability Index – DSI) and substrate (Substrate
Suitability Index – SSI) variables, according with Eq. (1):
HSI=VSI×DSI×SSI.
The product of the HSI by the influencing area (A) of the corresponding model
ith node defines the weighted usable area (WUA) of that node. The sum of
the WUA result in the total amount of habitat suitability for the study site,
as described by Eq. (2), is
WUA=∑n=1iAi×HSIi=f(Q).
Considering that the BACI approach (before-after control-impact) is generally
the best way of detecting impacts or beneficial outcomes in river systems
(Downes et al., 2002), the resulting WUAs were then compared to the natural
habitat in a census-based benchmark. The equality of proportions between
habitat availabilities was tested using the χ2 test for proportions
in the R environment, while deviations were measured using the most commonly
used measures of forecast accuracy, namely, root mean square deviation
(RMSD), mean absolute deviation (MAD), and mean absolute percentage deviation
(MAPD). In all cases, smaller values of these measures indicate better
performance in parameter estimation.
Workflow of the modeling procedure
The workflow of the modeling procedure is presented in Fig. 3. Firstly, the
calibrated version of the riparian vegetation model is used to produce the
riparian landscape scenarios according to each of the considered flow
regimes. In each modeling run, this model uses as inputs one of the specific
flow regimes mentioned and models the effects of a decade of such a flow
regime in the local riparian vegetation. The output of the model is an
expected riparian vegetation landscape map (detailed by succession phases)
resulting from the inputted flow regime. This map is converted into a channel
roughness map by attributing to each riparian succession phase a specific
effective roughness height based on the expert knowledge of the authors, on
the literature (e.g., Barnes, 1967; Chow, 1959; Fisher and Dawson, 2003), and
on the calibration results of the models. The considered roughness values of
each succession phase are provided as the Supplement (Sect. S1 – Tables S3
and S4). These roughness maps are one of the inputs of the River2D model.
Methodological scheme representing the workflow of the modeling
procedure. White arrows stand for direct inputs, striped white arrows for
model outputs, and grey arrows for variable conversion processes.
Secondly, the River2D hydrodynamic model is used to determine the
water depths and flow velocities at the microhabitat scale (already
considering each of the roughness maps coming from the conversion of the
CASiMiR-vegetation output vegetation maps) and to compute the weighted usable
areas of the considered fish species using the previous calculated variables
and the inputted information regarding the observed fish species habitat
preferences for water depth and flow velocity. This is done similarly using
each of the riparian landscape scenarios. For each scenario run, the outcome
of this model is therefore the weighted usable area of each of the considered
species and life stages for each of the discharges considered in the Eflow
regime.
Expected patch mosaic of the riparian vegetation habitats shaped by
the natural, Eflow, and Eflow&Flush flow regimes (detailed by succession
phase, namely, initial phase – IP, pioneer phase – PP, early succession
woodland phase – ES, established forest phase – EF, and mature forest phase
– MF) in the OCBA study site (a) and in the OCPR study
site (b).
ResultsRiparian vegetation modeling
Different riparian landscapes resulted from the riparian vegetation modeling
according to the considered flow regimes in both case studies (Fig. 4).
Nonetheless, the modeled response of riparian vegetation to each flow regime
is similar in the two study sites. The riparian landscape, driven by the
natural flow regime, presents a river channel that is largely devegetated,
where the Initial (IP) and Pioneer (PP) phases together represent
approximately 43 and 35 % of the study site areas in OCBA and OCPR,
respectively. In this riparian landscape, the Early Succession Woodland phase
(ES) can only settle in approximately 8 % of OCBA and 1 % of OCPR
areas. The floodplain succession phases, namely, the Established Forest phase
(EF) and Mature Forest phase (MF), represent nearly 40 and 10 % of the
study area for OCBA, and close to 42 and 23 % for OCPR, respectively.
In contrast, the riparian landscape created by the Eflow regime is where the
riparian vegetation encroachment is more prominent. Herein, riparian
vegetation settles in the channel and evolves toward mature phases due to the
lack of the river flood disturbance. IP is now reduced to approximately
3 % in OCBA and 6 % in OCPR, while PP is non-existent in both cases.
ES covers up to approximately 48 and 26 % of the corresponding study
areas, whereas EF and MF maintain about the same area in both case studies.
The riparian landscape driven by the Eflow&Flush regime shows the capacity
of this flow regime to hold back vegetation encroachment in both cases. In
this riparian landscape scenario, IP and PP are maintained at approximately
30 % of the study site area in both case studies, whereas ES is kept
under 21 % in OCBA and only 2 % in OCPR. Once again, EF and MF
preserve their areas in both case studies.
Summing up, the results of the riparian vegetation modeling show a riparian
landscape degradation by vegetation encroachment in the Eflow landscape
scenario when compared with the natural riparian landscape. Instead, the
Eflow&Flush landscape scenario keeps approximately the same patch
disposal and succession phase's proportion as the natural landscape and
therefore does not present evidence of riparian landscape degradation.
Hydraulic characterization of OCBA (a) and
OCPR (b) according to the different expected riparian vegetation
habitats driven by the Eflow, Eflow&Flush, and natural flow regimes (data
obtained from 2-D hydrodynamic modeling). Different letters stand for
statistically significant differences between groups (t-test). Boxplots
portray the non-outlier value range, thick black lines the median value, and
black dots the mean values.
Hydrodynamic modeling
The changes undertaken by the riparian vegetation facing different flow
regimes are able to modify the hydraulic characteristics of the river
stretches (Fig. 5). Channel effective roughness heights (ks) change
dramatically according to the considered riparian landscapes, increasing
proportionally to the encroachment level of vegetation in the study sites. In
both case studies, the ks values of the Eflow landscape are clearly
distinct and higher compared to the other two riparian landscapes (Fig. 5).
The ks values in the Eflow&Flush landscape were found to be between
the values of Eflow and natural landscapes in the case of OCBA, and were very
similar to the natural landscape in the case of OCPR (Fig. 5).
Notwithstanding this, in both case studies, the ks mean values are
statistical significantly different between all three riparian landscapes
(test results in the Supplement; Sect. S3 – Table S8). The mean ks of
the Eflow, Eflow&Flush, and natural landscapes are 0.999, 0.709, and
0.462 m, respectively, in OCBA, and 1.034, 0.742, and 0.7178 m,
respectively, in OCPR.
Changes also occur in flow depth and flow velocity for the considered
discharge range of the proposed environmental flows (Fig. 5). Although not so
noticeable due to the great amount of data, differences are statistically
significant. In OCBA, the Eflow landscape creates a circumstance with
statistically significant higher depths (mean depth is 0.402 m) and lower
flow velocities (mean flow velocity is 0.128 m s-1) than the natural
and Eflow&Flush landscapes. The t-tests on water depths (H0: the true
difference in the means is equal to 0) revealed highly significant p-values
(< 0.001), respectively, for the comparisons between Eflow and
natural flow regimes, and Eflow and Eflow&Flush flow regimes. The
t-tests on flow velocities also derived a highly significant p-value
(< 0.001) in both the comparisons of natural versus Eflow regimes
and Eflow versus Eflow&Flush flow regimes (test results in the Supplement;
Sect. S3 – Tables S9 and S10). In contrast, depth and flow velocity are not
significantly distinguishable between the natural and Eflow&Flush
landscapes, where mean depth and flow velocity are 0.397 m and
0.136 m s-1, respectively, in the former, and 0.399 m and
0.135 m s-1, respectively, in the latter.
For the OCPR study site, flow depths are not significantly different
(t-tests obtained p-values of 0.122 for natural versus Eflow regimes and
0.098 for Eflow versus Eflow&Flush flow regimes). Mean values of flow
depth for Eflow, Eflow&Flush, and natural landscapes are 0.420, 0.417, and
0.418, respectively. Nonetheless, flow velocities are different with
statistical significance as the p-values of the t-tests for natural
versus Eflow and for Eflow versus Eflow&Flush were highly significant
(< 0.001). The Eflow landscape creates statistically significantly
lower flow velocities (0.271 m s-1) when compared to the statistically
significantly indistinct Eflow&Flush (0.277 m s-1) and natural
(0.278 m s-1) landscapes (test results in the Supplement; Sect. S3 –
Tables S9 and S10).
Furthermore, when comparing water depths and flow velocities point by point,
one can find differences between scenarios of up to 10 cm in water depth and
more than 40 cm s-1 in flow velocity. Accordingly, there are locations
where the considered hydraulic parameters change considerably, shifting the
habitat preference of fishes in one or two classes of the corresponding
habitat preference curves.
In general, the Eflow landscapes present an increased channel roughness
interfering with river flow and creating increased water depths and slower
flow velocities when compared with the natural landscape. By contrast,
despite the increased channel roughness of the Eflow&Flush landscape, the
water depths and flow velocities are very similar to the ones in the natural
landscape. These results demonstrate that an environmental flow addressing
exclusively fish requirements is not capable of preserving the habitat
availability of the aquatic species for which it was proposed in the long
term.
Fish-weighted usable areas provided by the fish-addressed
environmental flow regime (Eflow) flowing through the different riparian
landscape scenarios that originated from a decade of three different flow
regimes (natural, Eflow&Flush, and Eflow) at the OCBA (a) and
OCPR (b) study sites.
Analysis of the aquatic habitat suitability for fish species
During a hydrological year, each riparian landscape provides different WUAs
for the target fish species, with the same environmental flow regime
addressing fish species (Fig. 6). Differences
from the natural habitat suitability are greater in the Eflow landscape for
both case studies. In OCBA, major differences in the WUA can be found almost
all year round for the barbel juveniles, throughout autumn and winter months
for the nase juveniles and during spring months for the calandino. Compared
to the natural landscape, the WUA modifications instilled by the Eflow
landscape are on average approximately 12 %, and are higher than 17 % in
a quarter of the cases reaching 80 % in an extreme situation.
Particularly, the Eflow landscape provides less habitat suitability during
autumn and winter months for the barbel and nase juveniles, c. 17 and
14 %, respectively. Likewise, in this riparian landscape, the habitat
suitability during spring months increases approximately 23 % for the
barbel juveniles and approximately 20 and 27 % for the calandino juveniles
and adults, respectively. On the other hand, throughout the year, the
Eflow&Flush landscape provides a WUA very similar to the natural
landscape. The habitat changes created by the Eflow&Flush landscape are
on average approximately 2 % and never reach 8 % for all species and
life stages.
As for OCPR, major differences in WUA are seen almost all year round for
calandino and nase, and exist particularly in spring months for barbel. WUA
modifications due to the Eflow landscape are on average near 29 %, a quarter being more than 50 % and
reaching up to more than 100 % different in the most extreme case. The
Eflow landscape consistently provides less habitat suitability during the
autumn and winter months for the nase juveniles and adults, ca. 50 and
38 %, respectively, while the habitat suitability increases by
approximately 46 % in calandino. Moreover, the Eflow landscape provides
an increased WUA during spring months in approximately 18 % of the barbel
adults and 71 % of the calandino adults, while it decreases the habitat
on average for approximately 7 % of the remaining species and life
stages. Also in this case study, the Eflow&Flush landscape provides a WUA
very similar to the natural landscape throughout the year. The habitat
changes created by the Eflow&Flush landscape are on average near 3 %
and always less than 17 % for all species and life stages. Accordingly,
in both case studies, the WUA differences evidenced in the Eflow landscape
proved to be significant in several months by the χ2 test, whereas
this was never the case for the Eflow&Flush landscapes (test results
provided in the Supplement; Sect. S3 – Tables S11, S12, S13, and S14).
The riparian-induced modifications on the WUAs are also confirmed by all the
employed deviation measures (Table 2). According to RMSD, MAD, and MAPD, the
habitat provided by the Eflow landscape is always farther apart from the
natural habitat for all species and life stages. In OCBA, the larger
deviations occur for the barbel juveniles and nase adults, whereas in OCPR,
the calandino adults and the barbel juveniles are the ones enduring greater
habitat deviations from the natural circumstances. All together, these
results reveal that the disregard of
riparian requirements into environmental flows can derail the goals of
environmental flows addressing only aquatic species by an extent of
approximately an average of 12 to 29 % of the fish WUAs in the considered
study sites as a result of the riparian landscape degradation. On the other
hand, results reveal that environmental flows regarding riparian requirements
are able to maintain the habitat availability of fish species as the WUAs in
the study sites never change on average more the 3 % in a decade.
Deviation analysis of the weighted usable areas for the considered
regulated flow regimes benchmarked by the natural flow regime (RMSD – root
mean square deviation, MAD – mean absolute deviation, MAPD – mean absolute
percentage deviation). Values stand for the habitat availability deviation,
in area and percentage, of the environmental flow regimes compared to the
natural habitat availability of each species and life stage.
OCBA study site OCPR study site Eflow Eflow&Flush Eflow Eflow&Flush RMSDMADMAPDRMSDMADMAPDRMSDMADMAPDRMSDMADMAPD(m2)(m2)(%)(m2)(m2)(%)(m2)(m2)(%)(m2)(m2)(%)Luciobarbus bocagei (juv.)86.0072.1015.4012.177.242.5226.2317.3735.552.511.500.63Luciobarbus bocagei (adult)29.4620.555.832.872.121.5512.947.7323.153.441.793.01Pseudochondrostoma polypepis (juv.)128.2186.1411.589.425.722.2645.4232.7134.431.550.922.51Pseudochondrostoma polypepis (adult)7.325.8518.702.171.372.109.007.0010.340.510.352.42Squalius alburnoides (juv.)44.0528.168.466.204.062.1033.1027.7828.372.441.352.18Squalius alburnoides (adult)92.4152.4710.237.495.312.3761.7647.8340.540.960.632.90Discussion
This study evaluated the benefits of incorporating riparian requirements
into environmental flows by estimating the expected repercussions of
riparian changes driven by regulated flow regimes on the fish long-term
habitat suitability. To this end, the riparian vegetation was modeled for
10-year periods according to three different flow regimes and results were
inputted as the habitat basis for the hydrodynamic modeling and subsequent
assessment of the fish habitat suitability in those riparian landscapes.
Such ecological modeling approach, where a joint analysis is performed while
embracing a suitable time response for the ecosystems involved, enables a
realistic biological-response modeling and substantiates the long-term
research that is required in environmental flow science (Arthington, 2015;
Petts, 2009). Furthermore, this approach allows one to foresee and assess
the outcome of recommended flow regimes, which is an essential topic but has
been poorly considered in environmental flow science (Davies et al., 2013;
Gippel, 2001). This research provides an insight of the expected long-term
effects of environmental flows in river ecosystems, therefore unveiling the
potential remarkable role of riparian vegetation on the support of
environmental flows efficiency, which can transform the actual paradigm in
environmental flow science.
During modeling, geomorphology was considered immutable and sediment
transport that originated from the environmental flow regimes was
disregarded. River morphodynamics and their interactions with riparian
vegetation constitute an important river process in many rivers, particularly
in fine sediment rivers (e.g., Corenblit et al., 2009, 2011; Gurnell et al.,
2012; Gurnell, 2014). However, the research on the temporal scales of
geomorphic and ecological processes is still scarce in coarse-bed rivers
(Corenblit et al., 2011), and simultaneously more complex and uncertain (Yasi
et al., 2013). The error predictions from the best hydraulic predictors in
this type of river can range from 50 to 200 % (Van Rijn, 1993; Yasi et
al., 2013). Disregarding such processes in these study sites was carefully
considered. Given the above and the arguments mentioned in the methods
section, we are confident that this option in this case will not bring
tangible shortcomings to this research. Furthermore, the possible riverbed
degradation effects due to the release of sediment-starving floods by the dam
were not tested because according to our expert knowledge this will not pose
a problem in this case. Such floods with similar recurrence intervals were
already tested by Rivaes et al. (2015) in two river stretches of much smaller
grain size (pebbles and sand) and results showed in both cases that such
flood discharges were not relevant for riverbed degradation. The influence of
fish species on geomorphology and riparian vegetation by ecosystem
engineering, as was mentioned in the introduction, was not considered either
during this study as it seemed fairly unrealistic in these case studies due
to the general dimension of riverbed particles.
The results of the vegetation modeling illustrate how the natural flow regime
generates morphodynamic disturbances, without which the riparian vegetation
is able to settle and age in the river channel. This is an important outcome
that is essential to remember when providing environmental flow instructions.
Subsequently, microhabitat analysis demonstrated that changes in the riparian
landscape induce modifications in the hydraulic characteristics of the river
stretches. The differences in the mean values of these parameters are subtle
between riparian landscapes, but are statistically significant. Furthermore,
a detailed analysis using a pairwise comparison of flow depths and velocities
between scenarios shows that modifications can reach 10 cm in water depth
and more than 40 cm s-1 in flow velocity in some places. The
hydrodynamic modeling results show that the water flowing near the margins is
more affected than the water flowing in deeper areas of the river channel.
One reason for these results is certainly because this study is about the
effects of riparian vegetation encroachment on the physical habitat due to
the colonization of the river margins by woody riparian vegetation.
Accordingly, there are locations where the considered hydraulic parameters
change considerably, shifting the habitat preference of fishes in one or two
classes of the corresponding habitat preference curves. These changes are particularly important considering that
an alteration of one class regarding these parameters is sufficient to change
fish preferences from near null to maximum and vice versa in many cases, as can
be seen in the preference curves provided in the Supplement (Sect. S3 –
Figs. S10, S11, and S12).
The hydrodynamic modeling also indicated changes directly affecting the
habitat suitability of the existing fish species according to the riparian
landscape. Through time, the riparian landscape shaped by the Eflow regime
diverged in habitat suitability from the natural and Eflow&Flush
landscapes, and there were cases where the habitat suitability was modified
by more than double. The relationship between fish assemblages and habitat
has long been acknowledged (e.g., Clark et al., 2008; Matthews, 1998; Pusey
et al., 1993) and can have a significant impact on the ecological status and
function of the existing fish communities (Freeman et al., 2001; Jones et
al., 1996; Randall and Minns, 2000). Effectively, habitat loss is the major
threat concerning fish population dynamics and biodiversity (Bunn and
Arthington, 2002), thereby promoting population changes with a proportional
response to the enforced habitat change (Cowley, 2008). This is particularly
true for the fish species considered in this study (Cabral et al., 2006). The
habitat decrease for barbel and nase during the autumn and winter months
jeopardizes this species survival by refuge loss, which is particularly
important in flashy rivers (Hershkovitz and Gasith, 2013), such as the Ocreza
River and Mediterranean rivers in general. On the other hand, the habitat
change during the spring months undermines the spawning activity and
consequently the sustainability of future population stocks
(Lobón-Cerviá and Fernandez-Delgado, 1984). The habitat increase in
calandino during this period can be ecologically tricky due to the habitat
plasticity of this species (Doadrio, 2011; Gomes-Ferreira et al., 2005), as
well as its characteristic adoption for an r-selection strategy as an
evolutionary response to frequently disturbed environments (Bernardo et al.,
2003). Above all, one should not ignore the fact that the relationships
between fish assemblages and habitat are extremely complex (e.g., Diana et
al., 2006; Hubert and Rahel, 1989; Santos et al., 2011), a consequence of the
actual natural conditions (Poff and Allan, 1995; Poff et al., 1997) that when
disrupted may allow the expansion of more generalist and opportunistic fauna
(Poff and Ward, 1989).
Our results indicate that environmental flows taking into account riparian
vegetation requirements are able to preserve the naturalness of the riparian
landscape and, consequently, the maintenance of the fish habitat suitability.
Accordingly, the implementation of such measures in place of using
environmental flows addressing only fish requirements can provide significant
positive ecological effects in downstream reaches (Lorenz et al., 2013; Pusey
and Arthington, 2003) and additional ecosystem services like stream bank
stability, flood risk reduction, or wildlife habitat
(Berges, 2009; Blackwell and Maltby, 2006) while imposing minor revenue
losses on dam managers (Rivaes et al., 2015).
The implementation of such environmental flows could provide an additional
way to attain the “good ecological status” required by the Water Framework
Directive (WFD). In addition, taking up a procedure such as this one can act
both as “win–win” and “no-regret” adaptation measures during the second
phase of the WFD, because it potentiates the improvement of other ecological
indicators and mitigates the impacts of flow regulation, while being robust
enough to account for different scenarios of climate change (EEA, 2005).
Water science still lacks strong links between flow restoration and its
ecological benefits (Miller et al., 2012), particularly regarding long-term
monitoring of environmental flow performance (King et al., 2015, and
citations therein). Nevertheless, the outcomes of this study are a product of
long-term simulations by models that were calibrated and validated for the
corresponding watershed with local data in natural river flow conditions.
This standard procedure in modeling strengthens confidence in our predictions
as the models proved to correctly replicate the response of the riparian and
fish communities when paralleled with simultaneous observational data. In
addition, model uncertainty due to estimation uncertainty in input parameters
was previously assessed by means of sensitivity analyses of both models. In
either case the models proved to be quite robust to the uncertainty of
estimated parameter inputs (see Rivaes et al., 2013, and Boavida et al.,
2013), which reveals a relatively small uncertainty in the
model outputs and provides
additional confidence in the results.
In conclusion, we predict a change in fish habitat suitability according to
the long-term structural adjustments that riparian landscapes endure
following river regulation. These changes can be attributed to the effects
that altered riparian landscapes have on the hydraulic characteristics of the
river stretches. In our view, environmental flow regimes considering only the
aquatic biota are expected to become obsolete in a few years due to the
alteration of the habitat premises on which they were based. This situation
points to the unsustainability of these environmental flows in the long term,
failing to achieve the desired effects on aquatic communities for which those
were proposed in the first place. An environmental flow regime that
simultaneously considers riparian vegetation requirements contributes to the
preservation of the hydraulic characteristics of the river channel at the
natural riverine habitat standards, thereby maintaining the habitat
assumptions that support the environmental flow regimes regarding aquatic
communities. Consequently, accounting for riparian vegetation requirements
poses an essential measure to ensure the effectiveness of environmental flow
regimes in the long-term perspective of the fluvial ecosystem.
Riverbed topography, hydraulic measurements, riparian
vegetation, and fish sampling were collected by the authors and are available
at 10.5281/zenodo.839531 (Rivaes et al., 2017). Both the River2D
and CASiMiR-vegetation models are freeware available at
http://www.river2d.ualberta.ca/download.htm and
http://www.casimir-software.de/ENG/download_eng.html, respectively.
The Supplement related to this article is available online at https://doi.org/10.5194/hess-21-5763-2017-supplement.
The authors declare that they have no conflict of interest.
Acknowledgements
This research was financially supported by the Fundação para a
Ciência e a Tecnologia (FCT) under project UID/AGR/00239/2013. Rui Rivaes
benefited from a PhD grant sponsored by the FCT (SFRH/BD/52515/2014). Isabel
Boavida was supported by a postdoctoral grant (SFRH/BPD/90832/2012) also
sponsored by the FCT. José Maria Santos was supported by a postdoctoral
grant from the MARS project (http://www.mars-project.eu). The
Portuguese Institute for Nature Conservation and Forests (ICNF) provided the
necessary fishing and handling permits. Edited
by: Stan Schymanski Reviewed by: Maurits Ertsen and one
anonymous referee
ReferencesAcreman, M.: Ethical aspects of water and ecosystems, Water Policy, 3,
257–265, 10.1016/S1366-7017(01)00009-5, 2001.Acreman, M. C. and Ferguson, J. D.: Environmental flows and the European
Water Framework Directive, Freshwater Biol., 55, 32–48, 10.1111/j.1365-2427.2009.02181.x, 2010.Acreman, M. C., Aldrick, J., Binnie, C., Black, A., Cowx, I., Dawson, H.,
Dunbar, M., Extence, C., Hannaford, J., Harby, A., Holmes, N., Jarritt, N.,
Old, G., Peirson, G., Webb, J., and Wood, P.: Environmental flows from dams:
the water framework directive, P. I. Civil Eng.-Eng. Su., 162, 13–22,
10.1680/ensu.2009.162.1.13, 2009.Acreman, M., Arthington, A. H., Colloff, M. J., Couch, C., Crossman, N. D.,
Dyer, F., Overton, I., Pollino, C. A., Stewardson, M. J., and Young, W.:
Environmental flows for natural, hybrid, and novel riverine ecosystems in a
changing world, Front. Ecol. Environ., 12, 466–473,
10.1890/130134, 2014.Alexander, J. and Cooker, M. J.: Moving boulders in flash floods and
estimating flow conditions using boulders in ancient deposits, Sedimentology,
63, 1582–1595, 10.1111/sed.12274, 2016.
Allan, J. D. and Castillo, M. M.: Stream Ecology: Structure and function of
running waters, Second edition ed., Springer, Dordrecht, NL, 436 pp., 2007.
Altman, D. G.: Practical Statistics for Medical Research, Chapman & Hall,
London, UK, 613 pp., 1991.
Arthington, A. H.: Environmental flows: saving rivers in the third
millennium, Freshwater Ecology Series, 4, Univ. of California Press, 406 pp.,
2012.Arthington, A. H.: Environmental flows: a scientific resource and policy
framework for river conservation and restoration, Aquat. Conserv., 25, 155–161, 10.1002/aqc.2560, 2015.
Arthington, A. H. and Zalucki, J. M.: Comparative Evaluation of
Environmental Flow Assessment Techniques: Review of Methods, in: Land and
Water Resources Research and Development Corporation, edited by: Arthington, A. H. and Zalucki, J. M., Canberra, AUSTRALIA,
141 pp., 1998.
Arthington, A. H., King, J. M., O'Keeffe, J. H., Bunn, S. E., Day, J. A.,
Pusey, B. J., Blüdhorn, D. R., and Tharme, R. E.: Development of an
holistic approach for assessing environmental water requirements of riverine
ecosystems, Proceedings of an International Seminar and Workshop on Water
Allocation for the Environment, Armidale, Australia, 1992, 69–76, 1992.
Arthington, A. H., Bunn, S. E., Poff, L. N., and Naiman, R. J.: The
challenge of providing environmental flow rules to sustain river ecosystems,
Ecol. Appl., 16, 1311–1318, 2006.Arthington, A. H., Naiman, R. J., McClain, M. E., and Nilsson, C.:
Preserving the biodiversity and ecological services of rivers: new
challenges and research opportunities, Freshwater Biol., 55, 1–16, 10.1111/j.1365-2427.2009.02340.x, 2010.
Barnes, H. H.: Roughness Characteristics of Natural Channels, Washington,
USA, 1967.
Benjankar, R., Egger, G., and Jorde, K.: Development of a dynamic floodplain
vegetation model for the Kootenai river, USA: concept and methodology, 7th
ISE and 8th HIC, 2009.Benjankar, R., Egger, G., Jorde, K., Goodwin, P., and Glenn, N. F.: Dynamic
floodplain vegetation model development for the Kootenai River, USA, J. Environ. Manage., 92, 3058–3070, 10.1016/j.jenvman.2011.07.017,
2011.Benjankar, R., Jorde, K., Yager, E. M., Egger, G., Goodwin, P., and Glenn,
N. F.: The impact of river modification and dam operation on floodplain
vegetation succession trends in the Kootenai River, USA, Ecol. Eng., 46, 88–97, 10.1016/j.ecoleng.2012.05.002,
2012.
Berges, S. A.: Ecosystem services of riparian areas: stream bank stability
and avian habitat, Master of Science, Iowa State University, Ames, Iowa,
USA, 106 pp., 2009.Bernardo, J. M., Ilhéu, M., Matono, P., and Costa, A. M.: Interannual
variation of fish assemblage structure in a Mediterranean river:
implications of streamflow on the dominance of native or exotic species,
River Res. Appl., 19, 521–532, 10.1002/rra.726, 2003.
Blackwell, M. S. A. and Maltby, E.: How to use floodplains for flood risk
reduction, European Communities, Luxembourg, Belgium, 144 pp., 2006.Boavida, I., Santos, J., Cortes, R., Pinheiro, A., and Ferreira, M.:
Assessment of instream structures for habitat improvement for two critically
endangered fish species, Aquat. Ecol., 45, 113–124, 10.1007/s10452-010-9340-x, 2011.Boavida, I., Santos, J. M., Katopodis, C., Ferreira, M. T., and Pinheiro,
A.: Uncertainty in predicting the fish-response to two-dimensional habitat
modeling using field data, River Res. Appl., 29, 1164–1174, 10.1002/rra.2603, 2013.Boavida, I., Santos, J. M., Ferreira, M. T., and Pinheiro, A. N.: Barbel
habitat alterations due to hydropeaking, J. Hydro-Environ. Res.,
9, 237–247, 10.1016/j.jher.2014.07.009, 2015.
Bovee, K. D.: A guide to stream habitat analysis using the Instream Flow
Incremental Methodology, U.S.D.I. Fish and Wildlife Service, Office of
Biological Services, Washington, 131 pp., 1982.
Bovee, K. D. and Milhous, R. T.: Hydraulic simulation in instream flow
studies: Theory and techniques. Instream Flow Information Paper: No. 5.
FWS/OBS-78/33, Fish and Wildlife Service, Fort Collins, Colorado, USA,
1978.Brisbane Declaration: The Brisbane Declaration. Environmental flows are
essential for freshwater ecosystem health and human well-being, Declaration
of the 10th International Riversymposium and International Environmental Flows
Conference, Brisbane, AUS, 2007, 1–7, 2007.Broadmeadow, S. and Nisbet, T. R.: The effects of riparian forest management
on the freshwater environment: a literature review of best management
practice, Hydrol. Earth Syst. Sci., 8, 286–305,
10.5194/hess-8-286-2004, 2004.Bunn, S. E. and Arthington, A. H.: Basic Principles and Ecological
Consequences of Altered Flow Regimes for Aquatic Biodiversity, Environ.
Manage., 30, 492–507, 10.1007/s00267-002-2737-0, 2002.
Cabral, M. J., Almeida, J., Almeida, P. R., Dellinger, T., Ferrand de
Almeida, N., Oliveira, M. E., Palmeirim, J. M., Queiroz, A. I., Rogado, L.,
and Santos-Reis, M.: Livro vermelho dos vertebrados de Portugal,
Instituto da Conservação da Natureza/Assírio & Alvim,
Lisboa, 660 pp., 2006.
Capon, S. J. and Dowe, J. L.: Diversity and dynamics of riparian
vegetation, in: Principles for riparian lands management, edited by: Lovett,
S. and Price, P., Land & Water Australia, Canberra, AUS, 3–33, 2007.Chase, J. W., Benoy, G. A., Hann, S. W. R., and Culp, J. M.: Small
differences in riparian vegetation significantly reduce land use impacts on
stream flow and water quality in small agricultural watersheds, J. Soil Water Conserv., 71, 194–205, 10.2489/jswc.71.3.194, 2016.
Chow, V. T.: Open channel hydraulics, McGraw-Hill, New York, USA, 680 pp.,
1959.Clark, J. S., Rizzo, D. M., Watzin, M. C., and Hession, W. C.: Spatial
distribution and geomorphic condition of fish habitat in streams: an
analysis using hydraulic modelling and geostatistics, River Res. Appl., 24, 885–899, 10.1002/rra.1085, 2008.
Clarke, A. O. and Hansen, C. L.: The Recurrence of Large Boulder Movement in
Small Watersheds of the Anza Borrego Desert, California, Yearbook of the
Association of Pacific Coast Geographers, 58, 28–61, 1996.Cohen, J.: A coefficient of agreement for nominal scales, Educ. Psychol. Meas., XX, 37–46, 10.1177/001316446002000104, 1960.Copp, G. H.: The habitat diversity and fish reproductive function of
floodplain ecosystems, Environ. Biol. Fish., 26, 1–27, 10.1007/bf00002472, 1989.Corenblit, D., Steiger, J., Gurnell, A. M., and Naiman, R. J.: Plants
intertwine fluvial landform dynamics with ecological succession and natural
selection: a niche construction perspective for riparian systems, Global Ecol. Biogeogr., 18, 507–520, 10.1111/j.1466-8238.2009.00461.x,
2009.Corenblit, D., Baas, A. C. W., Bornette, G., Darrozes, J., Delmotte, S.,
Francis, R. A., Gurnell, A. M., Julien, F., Naiman, R. J., and Steiger, J.:
Feedbacks between geomorphology and biota controlling Earth surface
processes and landforms: A review of foundation concepts and current
understandings, Earth-Sci. Rev., 106, 307–331,
10.1016/j.earscirev.2011.03.002, 2011.Cowley, D. E.: Estimating required habitat size for fish conservation in
streams, Aquat. Conserv., 18, 418–431, 10.1002/aqc.845, 2008.Curran, J. C. and Hession, W. C.: Vegetative impacts on hydraulics and
sediment processes across the fluvial system, J. Hydrol., 505,
364–376, 10.1016/j.jhydrol.2013.10.013, 2013.
Davis, R. and Hirji, R.: Environmental flows: concepts and methods. Water
Resources and Environment Technical Note no C1,
Environmental Flow Assessment series, World Bank, Washington, DC, 27 pp.,
2003.Davies, P. M., Naiman, R. J., Warfe, D. M., Pettit, N. E., Arthington, A.
H., and Bunn, S. E.: Flow-ecology relationships: closing the loop on
effective environmental flows, Mar. Freshwater Res., 65, 133–141,
10.1071/MF13110, 2013.
Diana, M., Allan, J. D., and Infante, D.: The influence of physical habitat
and land use on stream fish assemblages in southeastern Michigan, Am. Fish. S. S., 48, 359–374, 2006.
Doadrio, I.: Ictiofauna continental española: bases para su seguimiento,
Ministerio de Medio Ambiente y Medio Rural y Marino, Centro de
Publicaciones, Madrid, Spain, 2011.Dosskey, M. G., Vidon, P., Gurwick, N. P., Allan, C. J., Duval, T. P., and
Lowrance, R.: The Role of Riparian Vegetation in Protecting and Improving
Chemical Water Quality in Streams1, J. Am. Water Resour. As., 46, 261–277, 10.1111/j.1752-1688.2010.00419.x, 2010.
Downes, B. J., Barmuta, L. A., Fairweather, P. G., Faith, D. P., Keough, M.
J., Lake, P., Mapstone, B. D., and Quinn, G. P.: Monitoring ecological
impacts: concepts and practice in flowing waters, Cambridge University
Press, 434 pp., 2002.Dudgeon, D., Arthington, A. H., Gessner, M. O., Kawabata, Z. I., Knowler, D.
J., Lévêque, C., Naiman, R. J., Prieur-Richard, A. H., Soto, D.,
Stiassny, M. L. J., and Sullivan, C. A.: Freshwater biodiversity:
importance, threats, status and conservation challenges, Biol. Rev.,
81, 163–182, 10.1017/S1464793105006950, 2006.
Dyson, M., Bergkamp, G., and Scanion, J.: Flow. The Essentials of
Environmental Flows, in: IUCN, edited by: Dyson, M.,
Bergkamp, G., and
Scanion, J., Gland,
Switzerland and Cambridge, UK, 118 pp.,
2003.
EEA: Vulnerability and adaptation to climate change in Europe. Technical
report No. 7/2005, European Environment Agency, Copenhagen, DNK, 79 pp., 2005.
Encina, L., Granado-Lorencio, C. A., and Rodríguez Ruiz, A.: The
Iberian ichthyofauna: ecological contributions, Limnetica, 25, 349–368,
2006.
Ferreira, M. T., Pinheiro, A. N., Santos, J. M., Boavida, I., Rivaes, R.,
and Branco, P.: Determinação de um regime de caudais ecológicos
a jusante do empreendimento de Alvito, Instituto Superior de Agronomia,
Universidade de Lisboa, Lisboa, 136 pp., 2014.Fetherston, K. L., Naiman, R. J., and Bilby, R. E.: Large woody debris,
physical process, and riparian forest development in montane river networks
of the Pacific Northwest, Geomorphology, 13, 133–144, 10.1016/0169-555X(95)00033-2, 1995.
Fisher, K. and Dawson, H.: Reducing uncertainty in river flood conveyance –
roughness review, Department for Environment, Food & Rural Affairs,
Environment Agency, Lincoln, UK, 209 pp., 2003.FitzHugh, T. W. and Vogel, R. M.: The impact of dams on flood flows in the
United States, River Res. Appl., 27, 1192–1215, 10.1002/rra.1417, 2010.Freeman, M. C., Bowen, Z. H., Bovee, K. D., and Irwin, E. R.: Flow and
Habitat Effects on Juvenile Fish Abundance in Natural and Altered Flow
Regimes, Ecol. Appl., 11, 179–190, 10.2307/3061065, 2001.
Frissell, C., Liss, W., Warren, C., and Hurley, M.: A hierarchical framework
for stream habitat classification: Viewing streams in a watershed context,
Environ. Manage., 10, 199–214, 1986.García-Arias, A., Francés, F., Ferreira, T., Egger, G.,
Martínez-Capel, F., Garófano-Gómez, V.,
Andrés-Doménech, I., Politti, E., Rivaes, R., and
Rodríguez-González, P. M.: Implementing a dynamic riparian
vegetation model in three European river systems, Ecohydrology, 6, 635–651, 10.1002/eco.1331, 2013.
Gasith, A. and Resh, V. H.: Streams in Mediterranean Climate Regions:
abiotic influences and biotic responses to predictable seasonal events,
Annu. Rev. Ecol. Syst., 30, 51–81, 1999.Gillespie, B. R., Desmet, S., Kay, P., Tillotson, M. R., and Brown, L. E.: A
critical analysis of regulated river ecosystem responses to managed
environmental flows from reservoirs, Freshwater Biol., 60, 410–425, 10.1111/fwb.12506, 2014.
Gippel, C.: Australia's environmental flow initiative: Filling some
knowledge gaps and exposing others, Water Sci. Technol., 43,
73–88, 2001.Gomes-Ferreira, A., Ribeiro, F., Moreira da Costa, L., Cowx, I. G., and
Collares-Pereira, M. J.: Variability in diet and foraging behaviour between
sexes and ploidy forms of the hybridogenetic Squalius alburnoides complex
(Cyprinidae) in the Guadiana River basin, Portugal, J. Fish Biol.,
66, 454–467, 10.1111/j.0022-1112.2005.00611.x, 2005.Gorman, O. T. and Karr, J. R.: Habitat Structure and Stream Fish
Communities, Ecology, 59, 507–515, 10.2307/1936581, 1978.Gregory, S. V., Swanson, F. J., McKee, W. A., and Cummins, K. W.: An
Ecosystem Perspective of Riparian Zones: Focus on links between land and
water, Bioscience, 41, 540–551, 10.2307/1311607, 1991.Gurnell, A.: Plants as river system engineers, Earth Surf. Proc. Land., 39, 4–25, 10.1002/esp.3397, 2014.Gurnell, A. M., Bertoldi, W., and Corenblit, D.: Changing river channels:
The roles of hydrological processes, plants and pioneer fluvial landforms in
humid temperate, mixed load, gravel bed rivers, Earth-Sci. Rev., 111,
129–141, 10.1016/j.earscirev.2011.11.005, 2012.Hassan, M. A., Gottesfeld, A. S., Montgomery, D. R., Tunnicliffe, J. F.,
Clarke, G. K. C., Wynn, G., Jones-Cox, H., Poirier, R., MacIsaac, E.,
Herunter, H., and Macdonald, S. J.: Salmon-driven bed load transport and bed
morphology in mountain streams, Geophys. Res. Lett., 35, L04405, 10.1029/2007GL032997, 2008.Hershkovitz, Y. and Gasith, A.: Resistance, resilience, and community
dynamics in mediterranean-climate streams, Hydrobiologia, 719, 59–75, 10.1007/s10750-012-1387-3, 2013.
Hirji, R. and Davis, R.: Environmental Flows in Water Resources Policies,
Plans, and Projects, Environment and Development, The World Bank C1 –
Findings and Recommendations, 189 pp., 2009.
Hjulström, F. H.: Transportation of detritus by moving water: Part 1.
Transportation, in: Sp 10: Recent Marine Sediments, 5–31, 1939.Hubert, W. A. and Rahel, F. J.: Relations of Physical Habitat to Abundance
of Four Nongame Fishes in High-Plains Streams: A Test of Habitat Suitability
Index Models, N. Am. J. Fish. Manage., 9, 332–340,
10.1577/1548-8675(1989)009<0332:rophta>2.3.co;2, 1989.
Hughes, F. M. R.: Floodplain biogeomorphology, Prog. Phys. Geog., 21, 501–529, 1997.Hughes, F. M. R. and Rood, S. B.: Allocation of River Flows for Restoration
of Floodplain Forest Ecosystems: A Review of Approaches and Their
Applicability in Europe, Environ. Manage., 32, 12–33, 10.1007/s00267-003-2834-8, 2003.Jalón, D. G. D. and Gortázar, J.: Evaluation of instream habitat
enhancement options using fish habitat simulations: case-studies in the
river Pas (Spain), Aquat. Ecol., 41, 461–474, 10.1007/s10452-006-9030-x,
2007.
Jones, M. L., Randall, R. G., Hayes, D., Dunlop, W., Imhof, J., Lacroix, G.,
and Ward, N. J. R.: Assessing the ecological effects of habitat change:
moving beyond productive capacity, Can. J. Fish. Aquat. Sci., 53, 446–457, 1996.
King, J. M. and Tharme, R. E.: Assessment of the Instream Flow Incremental
Methodology and Initial Development of Alternative Instream Flow
Methodologies for South Africa, South African Water Research Commission,
604 pp., 1994.King, J. and Louw, D.: Instream flow assessments for regulated rivers in
South Africa using the Building Block Methodology, Aquat. Ecosyst. Health, 1, 109–124, 10.1016/S1463-4988(98)00018-9, 1998.King, A., Gawne, B., Beesley, L., Koehn, J., Nielsen, D., and Price, A.:
Improving Ecological Response Monitoring of Environmental Flows,
Environ. Manage., 55, 991–1005, 10.1007/s00267-015-0456-6, 2015.
Linnansaari, T., Monk, W. A., Baird, D. J., and Curry, R. A.: Review of
approaches and methods to assess Environmental Flows across Canada and
internationally. Research Document 2012/039, Canadian Science Advirosy
Secretariat (CSAS), 75 pp., 2012.
Lobón-Cerviá, J. and Fernandez-Delgado, C.: On the biology of the
barbel (Barbus barbus bocagei) in the Jarama River, Folia zoologica, 33,
371–384, 1984.Lorenz, A. W., Stoll, S., Sundermann, A., and Haase, P.: Do adult and YOY
fish benefit from river restoration measures?, Ecol. Eng., 61,
174–181, 10.1016/j.ecoleng.2013.09.027, 2013.Lytle, D. A. and Poff, N. L.: Adaptation to natural flow regimes, Trends Ecol. Evol., 19, 94–100,
10.1016/j.tree.2003.10.002, 2004.Maheshwari, B. L., Walker, K. F., and McMahon, T. A.: Effects of regulation
on the flow regime of the river Murray, Australia, Regul. River., 10, 15–38, 10.1002/rrr.3450100103, 1995.
Mäkinen, H. and Vanninen, P.: Effect of sample selection on the
environmental signal derived from tree-ring series, Forest Ecol. Manag., 113, 83–89, 1999.
Matthews, W. J.: Patterns in freshwater fish ecology, Springer Science &
Business Media, Norman, Oklahoma, USA, 756 pp., 1998.Merritt, D. M., Scott, M. L., Poff, L. N., Auble, G. T., and Lytle, D. A.:
Theory, methods and tools for determining environmental flows for riparian
vegetation: riparian vegetation-flow response guilds, Freshwater Biol.,
55, 206–225, 10.1111/j.1365-2427.2009.02206.x, 2010.Miller, K. A., Webb, J. A., de Little, S. C., and Stewardson, M.: Will
environmental flows increase the abundance of native riparian vegetation on
lowland rivers? A systematic review protocol, Environmental Evidence, 1,
1–9, 10.1186/2047-2382-1-14, 2012.Miller, K. A., Webb, J. A., de Little, S. C., and Stewardson, M. J.:
Environmental Flows Can Reduce the Encroachment of Terrestrial Vegetation
into River Channels: A Systematic Literature Review, Environ. Manage., 52, 1202–1212, 10.1007/s00267-013-0147-0, 2013.
Naiman, R. J., Décamps, H., and McClain, M. E.: Riparia – Ecology,
conservation and management of streamside communities, Elsevier academic
press, London, UK, 430 pp., 2005.Nilsson, C. and Berggren, K.: Alterations of Riparian Ecosystems Caused by
River Regulation, Bioscience, 50, 783–792, 10.1641/0006-3568(2000)050[0783:aorecb]2.0.co;2, 2000.
NRC, N. R. C.: Riparian Areas: Functions and Strategies for Management, The
National Academies Press, Washington, DC, USA, 444 pp., 2002.Palmer, M. A., Hondula, K. L., and Koch, B. J.: Ecological Restoration of
Streams and Rivers: Shifting Strategies and Shifting Goals, Annu. Rev. Ecol. Evol. S., 45, 247–269,
10.1146/annurev-ecolsys-120213-091935, 2014.Parasiewicz, P.: Using MesoHABSIM to develop reference habitat template and
ecological management scenarios, River Res. Appl., 23,
924–932, 10.1002/rra.1044, 2007.Petts, G.: Instream flow science for sustainable river management, J. Am. Water Resour. As., 45, 1071–1086, 10.1111/j.1752-1688.2009.00360.x, 2009.Poff, N. L. and Allan, J. D.: Functional Organization of Stream Fish
Assemblages in Relation to Hydrological Variability, Ecology, 76, 606–627, 10.2307/1941217, 1995.
Poff, N. L. and Ward, J. V.: Implications of streamflow variability and
predictability for lotic community structure: a regional analysis of
streamflow patterns, Can. J. Fish. Aquat. Sci., 46,
1805–1818, 1989.
Poff, L. N., Allan, J. D., Bain, M. B., Karr, J. R., Prestegaard, K. L.,
Richter, B. D., Sparks, R. E., and Stromberg, J. C.: The natural flow
regime, Bioscience, 47, 769–784, 1997.
Politti, E. and Egger, G.: Casimir Vegetation Manual, Environmental
consulting Ltd, Klagenfurt, AT, 76 pp., 2011.Politti, E., Egger, G., Angermann, K., Rivaes, R., Blamauer, B., Klösch,
M., Tritthart, M., and Habersack, H.: Evaluating climate change impacts on
Alpine floodplain vegetation, Hydrobiologia, 737, 225–243, 10.1007/s10750-013-1801-5, 2014.
Postel, S. and Richter, B.: Rivers for life: managing water for people and
nature, Island Press, Washington DC, USA, 243 pp., 2003.Pusey, B. J. and Arthington, A. H.: Importance of the riparian zone to the
conservation and management of freshwater fish: a review, Mar. Freshwater Res., 54, 1–16, 10.1071/MF02041, 2003.Pusey, B. J., Arthington, A. H., and Read, M. G.: Spatial and temporal
variation in fish assemblage structure in the Mary River, south-eastern
Queensland: the influence of habitat structure, Environ. Biol. Fish., 37, 355–380, 10.1007/bf00005204, 1993.
R Development Core Team: R: A language and environment for statistical
computing, R Foundation for Statistical Computing, Vienna, AT, 2011.Randall, R. G. and Minns, C. K.: Use of fish production per unit biomass
ratios for measuring the productive capacity of fish habitats, Can. J. Fish. Aquat. Sci., 57, 1657–1667, 10.1139/cjfas-57-8-1657, 2000.
Revenga, C., Brunner, J., Henninger, N., Kassem, K., and Payne, R.: Pilot
Analysis of Global Ecossystems: Freshwater Systems, World Resources
Institute, Washington, DC, 80 pp., 2000.
Richter, B. D. and Thomas, G. A.: Restoring environmental flows by
modifying dam operations, Ecol. Soc., 12, 1–12, 2007.Rivaes, R., Rodríguez-González, P. M., Albuquerque, A., Pinheiro,
A. N., Egger, G., and Ferreira, M. T.: Riparian vegetation responses to
altered flow regimes driven by climate change in Mediterranean rivers,
Ecohydrology, 6, 413–424, 10.1002/eco.1287, 2013.Rivaes, R., Rodríguez-González, P. M., Albuquerque, A., Pinheiro,
A. N., Egger, G., and Ferreira, M. T.: Reducing river regulation effects on
riparian vegetation using flushing flow regimes, Ecol. Eng., 81,
428–438, 10.1016/j.ecoleng.2015.04.059, 2015.Rivaes, R., Boavida, I., Santos, J. M., Pinheiro, A. N., and Ferreira, M. T.:
Data availability to ensure the reproducibility of the results of Rivaes et
al. (2017) in the journal HESSD,
10.5281/zenodo.839531, last access: 6 August 2017.Rood, S. B., Bigelow, S. G., Polzin, M. L., Gill, K. M., and Coburn, C. A.:
Biological bank protection: trees are more effective than grasses at
resisting erosion from major river floods, Ecohydrology, 8, 772–779, 10.1002/eco.1544, 2015.Ryan, D. K., Yearsley, J. M., and Kelly-Quinn, M.: Quantifying the effect of
semi-natural riparian cover on stream temperatures: implications for
salmonid habitat management, Fisheries Manag. Ecol., 20, 494–507, 10.1111/fme.12038, 2013.Salemi, L. F., Groppo, J. D., Trevisan, R., Marcos de Moraes, J., de Paula
Lima, W., and Martinelli, L. A.: Riparian vegetation and water yield: A
synthesis, J. Hydrol., 454, 195–202,
10.1016/j.jhydrol.2012.05.061, 2012.Santos, J. M., Ferreira, M. T., Godinho, F. N., and Bochechas, J.: Efficacy
of a nature-like bypass channel in a Portuguese lowland river, J. Appl. Ichthyol., 21, 381–388, 10.1111/j.1439-0426.2005.00616.x, 2005.Santos, J. M., Reino, L., Porto, M., Oliveira, J. O., Pinheiro, P., Almeida,
P., Cortes, R., and Ferreira, M.: Complex size-dependent habitat
associations in potamodromous fish species, Aquat. Sci., 73, 233–245, 10.1007/s00027-010-0172-5, 2011.
SNIRH: National Water Resources Information System, Instituto da
Água, I. P. (INAG), 2010.Statzner, B.: Geomorphological implications of engineering bed sediments by
lotic animals, Geomorphology, 157–158, 49–65,
10.1016/j.geomorph.2011.03.022, 2012.Statzner, B., Sagnes, P., Champagne, J.-Y., and Viboud, S.: Contribution of
benthic fish to the patch dynamics of gravel and sand transport in streams,
Water Resour. Res., 39, 1309, 10.1029/2003WR002270, 2003.
Steffler, P., Ghanem, A., Blackburn, J., and Yang, Z.: River2D, University of
Alberta, Alberta, CANADA, 2002.Stromberg, J. C., Tluczek, M. G. F., Hazelton, A. F., and Ajami, H.: A
century of riparian forest expansion following extreme disturbance:
Spatio-temporal change in Populus/Salix/Tamarix forests along the Upper San
Pedro River, Arizona, USA, Forest Ecol. Manag., 259, 1181–1198, 10.1016/j.foreco.2010.01.005, 2010.
Tabacchi, E., Lambs, L., Guilloy, H., Planty-Tabacchi, A.-M., Muller, E.,
and Décamps, H.: Impacts of riparian vegetation on hydrological
processes, Hydrol. Process., 14, 2959–2976, 2000.Tharme, R. E.: A global perspective on environmental flow assessment:
emerging trends in the development and application of environmental flow
methodologies for rivers, River Res. Appl., 19, 397–441, 10.1002/rra.736, 2003.
Thoms, M. C. and Parsons, M.: Eco-geomorphology: an interdisciplinary
approach to river science, The Structure and Management Implications of
Fluvial Sedimentary Systems, Alice Springs, Australia, 2002, IAHS Publ. no.
276, 113–119, 2002.
Thorp, J. H., Thoms, M. C., and Delong, M. D.: The Riverine Ecosystem
Synthesis. Toward Conceptual Cohesiveness in River Science, Elsevier,
London, UK, 208 pp., 2008.Uddin, F. M. J., Asaeda, T., and Rashid, M. H.: Factors affecting the
changes of downstream forestation in the South American river channels,
Environment and Pollution, 3, 24–40, 10.5539/ep.v3n4p24, 2014a.Uddin, F. M. J., Asaeda, T., and Rashid, M. H.: Large-Scale Changes of the
Forestation in River Channel Below the Dams in Southern African Rivers:
Assessment Using the Google Earth Images, Pol. J. Ecol., 62,
607–624, 10.3161/104.062.0407, 2014b.Van Looy, K., Tormos, T., Ferréol, M., Villeneuve, B., Valette, L.,
Chandesris, A., Bougon, N., Oraison, F., and Souchon, Y.: Benefits of
riparian forest for the aquatic ecosystem assessed at a large geographic
scale, Knowl. Managt. Aquatic Ecosyst., 408, 1–16, 10.1051/kmae/2013041, 2013.
Van Rijn, L. C.: Principles of sediment transport in rivers, estuaries and
coastal seas, Aqua Publications, Delft, NLD, 1993.
Viera, A. J. and Garrett, J. M.: Understanding interobserver agreement: the
Kappa statistic, Fam. Med., 37, 360–363, 2005.
Wainwright, J. and Mulligan, M.: Environmental Modelling: Finding
Simplicity in Complexity, John Wiley & Sons, Ltd, London, UK, 430 pp.,
2004.Wootton, J. T.: River Food Web Response to Large-Scale Riparian Zone
Manipulations, PLOS ONE, 7, e51839, 10.1371/journal.pone.0051839, 2012.
Wu, R. and Mao, C.: The assessment of river ecology and habitat using a
two-dimensional hydrodynamic and habitat model, J. Mar. Sci. Technol., 15, 322–330, 2007.
Yasi, M., Hamzepouri, R., and Yasi, A. R.: Uncertainties in Evaluation of
the Sediment Transport Rates in Typical Coarse-Bed Rivers in Iran, Journal of Water Sciences Research, 5, 1–12, 2013.