Journal of Environmental Management 119 (2013) 1e12

Contents lists available at SciVerse ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Perception of braided river landscapes: Implications for public

participation and sustainable management
Yves-Franois Le Lay*, Herv Pigay, Anne Rivire-Honegger
Universit de Lyon, CNRS-UMR 5600 Environnement, Ville, Socit, Site ENS de Lyon, 15 Parvis R. Descartes, 69362 Lyon Cedex 07, France

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 22 September 2011
Received in revised form
10 December 2012
Accepted 8 January 2013
Available online 19 February 2013

Over the past century, the ecologically-diverse, braided Magra River in Italy has narrowed, incised, and
lost many gravel bars due to the riparian vegetation encroachment following the decrease in bedload
supply and channel degradation. Motivated by the European Water Framework Directive, river scientists
and managers are beginning to plan projects to conserve and restore these dynamic mosaics of rare
habitats and processes. To support this objective, a study was conducted to assess how braided rivers are
perceived by different social groups in the area. In June, 2006, 127 people were surveyed using a photoquestionnaire consisting of ten photographs that depicted riverscapes with different proportions of
water, vegetation, and bed material. Respondents were asked to score each photograph in terms of
aesthetic value, benecial uses, and river management needs. Results showed that the photographs
depicting gravel bars were perceived as less aesthetically pleasing, so therefore they need an active
management. However, these perceptions differed amongst groups of participants, reecting their interests and objectives. This paper identies a distance between scientic and popular attitudes and
discusses implications for public participation, support for braided river restoration policy, and environmental education.
2013 Elsevier Ltd. All rights reserved.

Keywords:
Braided rivers
Environmental education
Environmental perception
Mediterranean mountain
Natural capital
Photo-questionnaire

1. Introduction
Braided rivers are characterized by numerous, mobile channels
that split and reconnect around gravel bars and islands colonized
by pioneer vegetation. Church and Jones (1992) identied three
preconditions that cause river braiding: an abundant sediment
supply; frequent and extreme variations in channel depth; and
non-cohesive, easily erodible riverbanks. In this bedload-rich river
type, gravel deposits can be several metres deep and 10e100 km
long (Gray and Harding, 2007). These are three-dimensional ecosystems, with water owing not only over the top of gravel beds,
but also horizontally and vertically through the gravels, often
reappearing as springs or spring-fed secondary channels (Brunke
and Gonser, 1997).
Spring-fed channels, abandoned oodplain channels, isolated
ponds and alluvial gravel bars together form a heterogeneous
mosaic of dynamic and connected riparian, hyporheic and
aquatic habitats, which in turn support a rich biodiversity of plant

* Corresponding author. Tel.: 33 4 37 37; fax: 33 4 63 47.

E-mail addresses: [email protected], [email protected]
(Y.-F. Le Lay), [email protected] (H. Pigay), [email protected]
(A. Rivire-Honegger).
0301-4797/$ e see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jenvman.2013.01.006

communities, macro-invertebrates, sh, and birds (Tockner et al.,

2006). Although braided river oodplains exhibit extreme instability, biological communities can persist given the overall magnitude or relative proportion of each habitat type within the system.
As soon as a habitat feature is destroyed, a similar one is maintained
or formed elsewhere, allowing mobile species to subsist within the
larger river oodplain (Arscott et al., 2002). Floodplain instability
also gives rise to a range of habitats in multiple stages of succession,
each occupied by distinct ecological communities (Whited et al.,
2007).
Braiding, which characterizes numerous rivers in the intermountain plains and foothills of the European Alps, appears to be
a phenomenon in balance with climatic and erosive conditions,
which are now changing (Peiry and Bravard, 1989; Surian, 1999).
Climate change combined with human impacts are altering natural
sediment supplies and ooding processes, consequently reducing
the width of the active channel and concentrating streamows
within a single-bed channel (decreasing habitat diversity and natural ood control).
The present study was conducted in the Regional Natural Park
of Montemarcello-Magra (4320 ha, including 18 villages), located
in the southernmost area of La Spezia province near the Tuscan
border, in Italy (Fig. 1). Within this protected area, a natural section of the Magra River, the side-channels of its main tributary,

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Zeri

Pontremoli

Filattiera
Bagnone
Mulazzo

Comano
Villafranca

Trasana

Licciana Nardi

Fivizzano
Podenzana

Aulla

Casola

Vara

Bolano

er

Riv

S. Stefano Magra
Vezzano L.
Arcola

Sarzana
Evolution during the last decade
(Rinaldi et al., 2009)
Ameglia

Tendency to degradation
Tendency to aggradation
0

5 km

Fig. 1. Localization map of the studied Magra River reach between Pontremoli and the sea. The villages where the interviews were done are underlined.

the Vara River, and most of the Ligurian wetlands comprise the
uvial park of the MagraeVara, which provides resting and
nesting areas for migratory birds and important spawning and
rearing areas for sh species including Alosa fallax (twait shad),
Lampetra uviatilis (river lamprey) and Petromyzon marinus (sea
lamprey). Even though the meandering lower Magra River (Bassa
Val di Magra) contains relict braided channels, it exhibits symptoms of intense human exploitation. The river drains a catchment
of about 1700 km2, a mean annual discharge of 40 m3 s1 and
a mean of maximum annual daily discharge of 683 m3 s1. Similar
to numerous rivers with large sediment loads that drain the Alps
and the Apennines (Pigay et al., 2009; Surian and Rinaldi, 2003),
the Magra River has experienced geomorphic and biogeographic
changes that have affected its braiding characteristics, including
a narrowing of the active channel, the encroachment of riparian
vegetation, and the incision of the river bed (Simoncini, 2008). In
fact, the natural reforestation of the headwaters of the basin at
the end of 19th century and early 20th century is the primary
cause of the reduced sediment supply (Rinaldi et al., 2005a,b). The
construction of dams and weirs, as well as considerable gravel
mining, has added to ongoing uvial changes since the mid1900s.

Scientic interest in the increasingly fewer examples of braided

rivers has grown in Europe and North America (Sambrook Smith
et al., 2006), resulting in the implementation of many restoration
studies, most notably in Central Europe (Hohensinner et al., 2004).
For river management purposes and notably for river restoration,
the question of what landscape is desired is a really critical question
for which this survey of perception can be a valuable tool to completing the technical investment made in such operations. This
study of environmental evaluation was therefore designed to understand how local stakeholders perceive a river channel that is
rich in gravels in order to better evaluate whether they will support
proposed braided channel restoration projects or not, and to guide
future public outreach and environmental education programs.
Data were collected on perceived naturalness, benecial uses of the
river, and need for improvement. The two key objectives of this
article were to test the hypothesis that these values are shared by
both academic and popular cultures, and to evaluate the degree and
conditions under which riverside communities can support river
conservation and restoration projects on braided rivers characterized by very low ows and large gravel deposits. Can we expect
river landscape appraisals of scientists and managers to be in
accordance with local stakeholders and public perception and

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

opinion? Are residents aware of the natural capital coming from

braided rivers or not? Due to the present degree of environmental
alteration, the need for sustainable sediment management is urgent and strongly supported by the basin authority (Rinaldi and
Simoncini, 2006). The questions raised above are clearly presented here as innovative, sustainable solutions for preserving
gravel resources and braided river habitats (Rinaldi et al., 2009).

2. Study site and survey methods

2.1. Design of the study
This survey of the visual impact of low ows and gravel deposits
on peoples perception and evaluation of riverscape quality is based
on theoretical perspectives of human and landscape perceptual
interaction (Zube et al., 1982). To evaluate perceptions of braided
rivers across several stakeholder groups, we elicited individual
responses using a photo-questionnaire e a well-respected technique that has been used in landscape perception and evaluation
research since the 1970s. Although surrogates may provoke a perceptual distortion, persons responses to an actual physical setting
are strongly and positively correlated to responses based on
a comprehensive photograph of the same scene. A meta-analysis of
11 papers yielded a combined correlation of 0.86 between preferences obtained in situ and preferences obtained through photographs (Stamps, 1990, p. 907). Many studies based on photoquestionnaires have been applied to aquatic environments (Le
Lay et al., 2012; Bulut and Yilmaz, 2009; Ryan, 1998), as well as to
more specic topics such as water (Pger et al., 2010), wood in
rivers (Le Lay et al., 2008) and riparian buffers (Kenwick et al.,
2009). Using this method, the landscapes portrayed in the photographs serve as effective substitutes for reality (Hull and Stewart,
1992; Shuttleworth, 1980). Most importantly, a photoquestionnaire survey permits differences among groups of participants to be quantitatively tested (Buhyoff et al., 1983). This protocol remains effective even with a small number of respondents
(Schroeder, 1984). Stamps (1992) applied a bootstrap procedure on
a set of 65 answers from respondents. His analysis showed splitblock correlations of 0.84, 0.86 and 0.90, respectively with 15, 20
and 25 subjects, for the ratings data. Therefore, large respondent
samples would probably not, in general, be a cost-effective protocol
for person/environment research (Stamps, 1992, p. 222).
Besides the environment itself, landscape preferences derive
from a framework of values, beliefs and experiences. Many observer
characteristics may have an effect upon descriptive and evaluative
responses (e.g., personality, socio-economic attributes, profession
and experience in terms of environmental management, and familiarity with a particular environment or a type of environment).
In order to assess the possible inuence of river uses, environmental knowledge and familiarity with the river, we surveyed ve
distinct groups of participants (total n 127). These groups consisted of 28 children, 13 years old and attending high school near
the Magra River (in the town of Villafranca); 38 children, 13 years
old and attending high school far-away from the river (towns of
Bagnone and Filattiera); 20 people with a scientic understanding
of river ecosystem function (post-graduate students and academics
from the University of Florence who work on the Magra River); 23
local river managers working for the government basin authority;
and 18 other Magra River inhabitants. We interviewed the children
in their schools with the authorization of the headmasters and
teachers; the invited scientists during a workshop about Magra
River geomorphology; the river managers in their ofces; and the
riverside inhabitants at Villafranca, a town located on the left bank
of the Magra River (Fig. 1).

A lot of disagreement between actors is associated with the

perception of nature. Residents may perceive naturalness very
differently than the way experts do (Buijs, 2009; Wohl and Merritts,
2007). In Switzerland, Junker and Buchecker (2008) highlighted
that aesthetic preferences relate more positively to ecomorphological quality than expected, and that the publics aesthetic preferences are primarily inuenced by perceived naturalness (p. 141). A participatory approach to sustainably managing
rivers and streams relies on consensus to reconcile natural resource
usage with economic, ecological, recreational, and cultural objectives (Gardiner, 1994). However, such an approach raises important
questions regarding implementation (Junker and Buchecker, 2008;
Kenwick et al., 2009). For these reasons, ten photographs of the
river were shown to each participant depicting different proportions of water, vegetation, and mineral bar, and different sizes of
gravel (Fig. 2a). The respondents evaluated each of the riverscapes
with a mark on the visual analogue scale (VAS) shown in Fig. 2b (Le
Lay et al., 2012). It ranged from zero (the lowest level of agreement)
to ten (the strongest level of agreement) for three following attributes: aesthetic quality (Is the river aesthetically pleasing to you?),
benecial uses of the river (Do you think the river meets the people
who visit it?); and the need for active river management (Do you
think the river needs improvement?).
In addition, participants identied the categories of uses to
which the riverscape shown would be most suited (1 e no activity;
2 e shing; 3 e swimming; 4 e navigation, 5 e relaxation, 6 e
gravel mining, 7 e hydroelectric power supply, 8 e domestic/
agricultural water supply, 9 e nature conservation). They also
identied appropriate types of management activities for each
riverscape (six categories: 1 e no management, 2 e improved
landscaping, 3 e wildlife protection, 4 e bank stabilization, 5 e
channel cleaning, 6 e engineered structures). Respondents were
allowed to select multiple uses and management actions.
2.2. Data analysis
The proportional area of water, vegetation, and mineral bar was
manually measured within each photograph using a transparent
grid (Table 1). The grid was laid upon each photograph so that the
number of boxes (0.5 cm on each side) occupied by each area could
be counted and converted to the percentage of the total surface
area covered in the photograph (Cossin and Pigay, 2001). The
sediment size was also visually assessed and categorized into four
classes: 1 e absent, 2 e gravels, 3 e pebbles, and 4 e large boulders
following the size scale of Wentworth (1922).
We performed all statistical procedures with R, a program for
data analysis and graphics (Ihaka and Gentleman, 1996). As each
photograph was graded on a specic VAS, the scores can be averaged and compared in various ways. As a result, for each of the
three variables (aesthetic value, benecial uses, management
needs), we considered the average scores for photographs, taking
into account potential differences between the four bedload size
classes.
We computed correlation coefcients of Bravais-Pearson from
scores of aesthetic value, benecial uses, management needs and
proportions of landscape components (water, mineral bar, and
vegetation) to measure the strength of the associations between
these six quantitative variables. These coefcients also identied
the attributes of riverscape quality which determined the decisionmaking factor.
We arranged tables for the three perceived riverscape variables
(aesthetic value, benecial uses, management needs) using mean
VAS scores per group of participants for each photograph. Thus
each table contained 5 columns corresponding to the 5 groups of
respondents and 10 rows corresponding to 10 slides. To investigate

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Fig. 2. (a) Photographs shown to the 127 survey respondents; and (b) example of the visual analogue scale (VAS).

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Table 1
Surface area of each photograph (A through J) occupied by water, vegetation, and
sediments, and the visually assessed sediment size (0 e no sediment, 1 e gravel, 2 e
pebbles, and 3 e large boulders).
Photograph

Water (%)

Vegetation (%)

Sediment (%)

Sediment
size

A
B
C
D
E
F
G
H
I
J

42.25
0.53
21.93
26.20
1.07
24.33
60.16
15.51
12.03
59.36

45.99
8.02
5.35
40.91
37.17
10.70
15.51
24.60
28.61
9.89

0.53
52.41
34.22
23.80
61.50
33.96
2.14
39.57
32.62
11.76

0
1
1
2
3
2
1
3
1
1

3. Results
3.1. The perception of the inuence of mineral bars on the riverine
landscape
Across participants, there were clear tendencies towards positive preferences for landscapes that depicted high proportions of
water (or conversely, low proportions of mineral bars). For example, there were strong positive associations between the proportion
of water depicted and average responses for aesthetic value
(r 0.76; p < 0.015) and benecial use (r 0.87; p < 0.0015)
(Table 2). Six signicant negative correlations were also identied.
Management needs are negatively correlated to benecial use
satisfaction (r 0.94; p < 0.0001), aesthetic value (r 0.90;
p < 0.0004), and proportion of water (r 0.70; p < 0.025).
Likewise, there were negative associations between the proportion
of mineral bars and benecial use (r 0.80; p < 0.005) and aesthetic value (r 0.70; p < 0.025).
Relationships between the area occupied by mineral bars and
each survey response variable are illustrated in Fig. 3. River landscapes dominated by mineral bars (slides B and E) were recognized
as less aesthetic, less usable and motivated respondents to require
an improvement. Sediment size also inuenced landscape perception (Fig. 3). For a given proportion of mineral cover, images
with large-sized boulders (slides E and H) were perceived as more
attractive, more usable, and requiring less active management than
Table 2
Correlation matrix (calculated with average Bravais-Pearson coefcients) for the
survey response variables (aesthetic value, benecial uses, and needs for management action) and the landscape components (area occupied by water, vegetation,
and mineral bar). Signicant values with p < 0.05 are in bold type.

Aesthetics
Uses
Management
Water
Vegetation
Mineral bar

Management Water

0.964 L0.904
L0.944

0.759
0.868
L0.899

Vegetation Mineral
bar
0.155
0.022
0.069
0.101

other riverscapes with lower sediment size fractions. Inversely, the

scenes I and C (covered with gravels) but also A and partly J (with
no sediment) were perceived as less attractive, less usable, and
requiring more active management than the predicted value for
a given mineral area.
3.2. Types of riverine landscapes and sociocultural contrasts

the variation between groups of participants, we performed a normalized Principal Component Analysis (nPCA) on each of the three
quantitative attributes describing riverscape perception. We provided the inertia ratio of each the signicant factors to statistically
validate the procedure.
To evaluate the categorical variables, we created bar plots from
contingency tables to determine relationships between benecial
uses and management actions recommended by the different
groups of respondents.

Aesthetics Uses

L0.702
L0.805
0.827
L0.916
0.121

The three normalized Principle Components Analyses (PCA)

identied strong similarities among preferences of each group, but
also highlighted notable differences, in particular for river managers (Fig. 4).
The two rst axes (principal components or PCs) for aesthetic
quality accounted for 91% of the total variation (73.7% and 17.4%,
respectively, for axes F1 and F2, Fig. 4a). The rst axis (the dominant
PC) was strongly associated with the relative surface area occupied
by water and sediments (shown using pie charts for each photograph in Fig. 4a). Photographs D, F, J, and especially G, were perceived as the most aesthetically pleasing landscapes. These images
were characterized by abundant owing, deep water (slide G), or by
a main channel shallow enough so that blue water was transparent
along the edges of gravel bars (photographs D, F, and J). The less
aesthetically pleasing scenes (B, C, E and I) depicted landscapes
dominated by mineral bars. Photograph H, which showed a river
owing through a mostly mineral surface, received an intermediate
rating. The responses of four groups of participants strongly contributed to the rst axis: high school children both near and far
from the river, river inhabitants, and scientists (right side of Fig. 4a).
In contrast, the F2 axis largely accounted for the distinct answers
provided by river managers, who appeared to respond to the proportion of vegetation (Table 1). The most vegetated landscapes
(mainly A and I) had negative coordinates on the axis F2. They were
perceived as relatively less attractive for managers. Although the
inuence of alluvial vegetation on aesthetic scores for managers
appears important, water and sediment were the key landscape
components determining aesthetic quality across groups.
Similar results were found for the PCA of benecial use responses. The rst two PCs for river usability explained 84.2% (axis
F1) and 11.8% (axis F2) of the total variation (Fig. 4b, inset). On the
factorial maps, the responses for river use corresponded well with
those provided for scenic beauty. Landscapes that elicited a high
degree of benecial usability were also recognized as being aesthetically pleasing. But the landscape components (e.g., water,
vegetation or mineral features) contributed more to these results.
Landscapes of water (G, D, J and F) had positive coordinates on axis
F1, unlike landscapes with more sedimentary deposits. Again, the
F2 axis appeared to account for differences in benecial use responses among groups that were associated with riparian vegetation (Fig. 4b). In comparison to aesthetic value, the responses given
by managers still differed from those of other groups of participants
and thus appear to be again related to the proportion of vegetation.
The distant high school children have also an opinion which slightly
differs from the other groups notably on photo E.
The rst two axes of a PCA performed on data from the need for
active river management explained 91% of the total variance
(respectively, 69.7% and 21.5% for axes F1 and F2; Fig. 4c, inset).
According to the correlation circle, the axis F1 was associated with
relative proportions of water and sediment, while the F2 axis was
structured by the area occupied by vegetation in the photographs.
The highest scores were associated with the photographs showing
many alluvia (scenes B, I, E, and H). If the attitudes of scientists, local
residents, and high school children living near the Magra were very
similar, responses provided by the more distant high school children were again far more inuenced by the presence of sediment
accumulations, whereas the responses of managers were driven by

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Beneficial uses

Management needs

Average score
6
7
8

Average score
6
7
8

Average score
6
7
8

Aesthetic value
G

A
x

E
B

C
A

IC

10

20
30
40
50
60
Area covered by sediment (%)
x

G
0

10

20
30
40
50
60
Area covered by sediment (%)

10

20
30
40
50
60
Area covered by sediment (%)

Undetermined grain size

Gravels
Pebbles
Large boulders

Fig. 3. The inuence of the proportion and size class of bedload shown in the photographs on the three survey response variables.

the proportion of vegetation. The scenes that were considered as

less aesthetically pleasing and less likely to provide benecial river
uses (B, E and I) were also recognized as requiring more management interventions to improve the current situation.
The ve participant groups had varying distributions of responses in the landscape evaluation (Fig. 5). There was a clear
difference between the two groups of high school children. Those
living near the Magra were quite homogenous in opinion, particularly with regard to aesthetic value and benecial uses, while
distant high school children expressed contrasting responses with
no emergent consensus. Local residents and managers had similar
aesthetic and benecial use perceptions, providing very few low
scores. In this context, scientists were distinguished by providing
very few high scores. In terms of management needs, the managers,
scientists, and high school children near Magra shared bimodal
distributions, with many low scores and intermediate scores. Local
residents also provided responses showing a bimodal distribution,
but bimodality was less clear, with more frequent low scores and
high scores. It should also be noted that river managers believed
that they should limit intervention, and that they attributed lower
scores to the need for river management than the scientists.
3.3. Characterization of benecial uses and management needs
The characterization of benecial river uses by the different
respondents groups for each of the ten river landscapes was
associated with the presence of water and alluvial deposits (Fig. 6).
When photographs depicted landscapes with abundant sediment
(B, C, E, H and I), observers were less likely to recognize benecial
river uses. Those images were associated with higher scores for no
activity. However, photographs B, C, E, H, and I were scored as
being conducive to relaxation and gravel mining. In addition, the
scenes in which water was abundant and aggregated into a single
channel (including A and G) were scored higher for swimming and
even more for shing activities.
Groups of participants had different characterizations of the uses
for which the riverscapes were the best suited. Managers and scientists shared similar attitudes. More often than others, they selected
categories such as nature conservation, shing, and relaxation. The
latter appeared to be the activity of choice among Magra River users,
who identied this as a use for six photographs containing a wellextended mineral area (B, C, F, H, I, and J). The responses provided

by high school children differed in that those living near the Magra
considered it favourable to shing, relaxation, and nature conservation, while distant high school children selected those uses less often.
No action as a river management need was most often selected (Fig. 7). However most respondents chose measures to improve
landscape and protect wildlife for photographs B and I, which
depicted more alluvial deposits. In addition, channel cleaning was
more often proposed for scenes C, E, and H, which contained high
proportions of sediment (34%, 61%, and 39%, respectively).
There were also clear differences between groups of respondents. Managers were less eager to recommend management
actions and suggested relatively few measures. However, they were
more motivated to act in favour of landscaping and wildlife protection when assessing photographs showing a single-bed lentic
channel (scenes A, G, and I). Faced with scene E, which showed
coarser sediments, managers were more inclined to clean the
channel or take more drastic engineering measures. Scientists were
more concerned with channel sections with concave streambanks,
which are sensitive to erosion and may require work to stabilize the
banks (photograph A). While the managers and scientists had
common concerns about sh habitat and landscaping, their responses differed in terms of risk: the latter were sensitive to the
problem of undercutting banks while the former were rather wary
of oods and the roughness of the channel. In addition, children
from high schools far-away from the Magra River were the most
motivated to intervene to manage the river.
Concerning the photo I managers scored differently than the
other groups: they did not consider this riverscape to be signicantly
different in terms of benecial uses (useful for shing and nature
conservation). In terms of management actions, they recommended
specic operations on vegetation to improve the landscape similarly
with photo A but they did not refer to any risk management actions.
Concerning the photo E distant high school children scored differently than the other groups: they did not consider such riverscape
useful for nature conservation, which differs from local high school
children opinion, and their choice in terms of management actions is
very variable, compared to the other scenes.
4. Discussion
The public opinion on the aesthetics, benecial uses and management needs of river landscapes is based both on personal

Fig. 4. Results of the three normalized Principal Component Analyses (nPCA) conducted on (a) aesthetic value, (b) benecial uses, and (c) management needs. On factorial maps
F1  F2 (on the left side), the circular pie charts indicate the positions of the ten photographs submitted to the 127 survey participants and depict the proportion of water (white), of
vegetation (gray), and of sediment (black), for each scene. The shaded labels (A through J) also identify the bedload size classes in each photograph. The inset bar chart in the upper
left of each factorial map shows the principal component eigenvalues. On the right-hand side, the positions of the ve groups of respondents with respect to the rst two principal
component axes, F1 and F2, are shown on the circles of correlations.

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Managers

Scientists

River users

Distant high
school children

Managers

Scientists

River users

Distant high
school children

Nearby high
school children

8
6
4
2
0
Managers

Scientists

River users

Distant high
school children

Nearby high
school children

10

Management needs

10

Beneficial uses

10

Aesthetic value

Nearby high
school children

Fig. 5. Intra-group variability of the scores attributed to each of the survey response variables.

preferences of the observer and on the characteristics of the environment (Dakin, 2003; Lothian, 1999). The results described here
rene and extend the conclusions described in the scientic literature by landscapers, behavioural geographers, and environmental
psychologists.
4.1. Natural capital and ecosystem services: conrming the
perceived beauty of water
The area occupied by water in a photograph was positively
correlated with perceived beauty and benecial uses of the riverscape, and negatively correlated with management needs. The
desirability of water is indeed one of the most prominent archetypes. It asserts itself in a Mediterranean context, where most
participants know river discharges undergoing large uctuations
during the year and from one year to another. We have however to
put his nding into perspective. Aesthetic quality decreases during
both low and high water periods (Brown and Daniel, 1991). The
negative impact of intense ows can be explained by a reduction in
the complexity of the landscape: a high level of water tends to
reduce the frequency of wet/dry thresholds, causing an apparent
uniformity of supercial velocity and removing islands and bars. In
contrast, the negative effect of low water is likely due to a loss of
vitality and vibrancy, with stagnant waters replacing rough waters.
The ow of water varies also with the aesthetic, recreational, or
nancial value of a river landscape.
Consequently, there is an optimum of discharge for the visual
quality of riverscapes (Pger et al., 2010). The attractiveness of riverscapes depends on both their features and recognized social benets. Surface freshwater (rivers, wetlands, or lakes) can provide many
goods and services to human society, including both market goods,
like water supply for municipal, industrial and agricultural users as
well as non-market goods such as habitat for plant and animal life,
biodiversity, aesthetic beauty, and recreation (Loomis et al., 2000;
Petrosillo et al., 2013; Raymond et al., 2009; Wilson and Carpenter,
1999). Even though the Magra River drains a large intensely cultivated and exploited plain, its ecological value was recognized by
protected areas: the Parco Fluviale della Magra, established in 1982,
and then the Parco della Magra e di Montemarcello in 1995. Braided
rivers play a signicant role in the functions of natural capital, both in
running the key processes that support the ow of goods and services
and in maintaining functions of natural capital resilient to change
(Ekins et al., 2003, p. 170). In addition to regulation and habitat
functions, and related ecosystem services, the Magra River provides
life-fullling services through the provision of aesthetic beauty, cultural, intellectual and spiritual inspiration. This aesthetic benet elicits

and enhances not only near-water recreational activities (hiking,

picnicking, and relaxation) but also boating, shing, and swimming.
Our results also highlight the inuence of sediment size on the
public preference for different visual riverscapes. Large boulders
were more visually appealing than small pebbles and gravels.
Beyond the mere presence of water and sediment, the organization
of landscape components must be considered. Large boulders
increase the number of items independently perceived in a photograph and provide more precise organization. They make the river
landscape more legible and more coherent. They permit a quick
understanding of the scene and offer enough information to generate visual interest (Porteous, 1996).
4.2. A gap between experts and residents?
This survey conrms some of the contradictions between the
perceptions of river experts and others when valuing braided
rivers. The results show that riverscapes with greater percentages
of mineral bars were perceived as less attractive and did not meet
the river use expectations of residents, motivating them to recommend management actions. Braided river landscapes with large
alluvial deposits were associated with lower perceptions of river
usability and greater desire among participants to improve the
landscape and protect wildlife, including some suggestions to clean
channels and install engineering structures.
Similarly, in a study on the Roubion River, a left bank tributary of
the Rhne River, in France, the identication of the least appreciated landscape was highly consensual among non-expert respondents (Le Lay et al., 2012). For the least pleasing photograph,
participants denounced the lack of water, the appearance of a river
bed that had been bulldozed, and the presence of gravel. The
Roubion River was considered as a poorly maintained and disorderly desert. Although representative of braided rivers in the
Rhne corridor, that landscape did not t the public preference.
Surprisingly, the participant responses showed that Italian scientists and local residents shared some similar opinions. Despite
their knowledge of quality habitats, scientists did not score braided
rivers any more positively. The responses of river managers were
unique in that their aesthetic evaluations were less inuenced by
the proportions of water and sediment in the photographs. Their
motivation to recommend management action was more closely
linked to the aesthetic value of the riverscape. However, the attitudes of managers and scientists were similar when characterizing
possible human and environmental uses of the river. More than
other groups, their concerns focused on nature conservation and
landscape enhancement.

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Photograph B

0.0

0.0

0.2

0.4

Frequency
0.2 0.4

Photograph A

Photograph D

0.0

Photograph F

0.0

Photograph H

0.0

0.0

0.2

Photograph G

0.4

Frequency
0.2 0.4

0.0

0.2

Frequency
0.2 0.4

Photograph E

0.4

0.0

0.2

0.4

Frequency
0.2 0.4

Photograph C

Photograph J

s
er

ts

us

is
nt

er

ie

iv

Sc

iv

er

an

us

ag
er
D
s
sc ista
ho nt
ol hi
ch gh
ild
re
N
n
e
sc ar
ho by
ol h
ch igh
ild
re
n

0.0
s
er

ts
is
nt
ie
Sc

an

ag
er
s
i
s
sc ta
ho nt
ol hi
ch gh
ild
re
N
n
sc ear
ho by
ol h
ch igh
ild
re
n

0.0

0.2

0.4

Frequency
0.2 0.4

Photograph I

us
e
er
R
iv

Sc

ie
nt

is

rs

ts

D
s
sc ista
ho nt
ol hi
ch gh
ild
re
N
n
e
sc ar
ho by
ol h
ch igh
ild
re
n

an

ag

er

0.00

0.15

Total
No activity
Fishing
Swimming
Navigation
Relaxation
Gravel mining
Hydroelectric generation
Water withdrawal
Nature conservation

Fig. 6. Categorization of benecial uses recommended by survey respondents for each riverscape photograph.

To reverse the progressive disappearance of braided reaches in

the Magra River and its tributaries, current restoration projects aim
to: 1) maintain the supply, storage, and transport of uvial sediments; 2) limit channel dredging; and, 3) increase sediment
recharge via increased lateral erosion and/or tributary sediment
inputs (Rinaldi et al., 2009). Specically, to restore oodplain space
to which the river is adapted, the processes of erosion and deposition would be reactivated via articial sediment recharge and the
destruction of engineered in-stream structures.
According to our results, such actions are likely to impact river
users and residents who are generally more sensitive to changes in
their local lifestyle than to regional and national conservation
strategies.
Should we attempt to meet public demands? Because we
appreciate what we have been taught to appreciate, a focus on
collective culture inhibits the new and reinforces the elements that

contribute to forming the ideal river landscape. Previous ndings

reected the publics ideal river setting to be one with open deciduous forest, with a mixture of grass and plants, or, grass and trees
overhanging the bank of a mature river, of natural form and without
an overabundance of vegetation in the water (House and Sangster,
1991, p. 315). Therefore a strategy of management or restoration
promoting a bedload-rich river type, with abundant gravel deposits,
unstable banks, many secondary and abandoned channels in a large
oodplain appear to be far from the public expectation.
4.3. Isnt there a way of reconciling the different points of view?
The gap between the perceptions and attitudes of professionals
and civil servants and those of the public can be harmful when
ignored (Dearden, 1981; Junker and Buchecker, 2008). However
simply recognizing this gap does not resolve all issues. How can we

0.4
0.0
0.8

0.8

0.4

Photograph D

0.8

0.0

0.4
0.8

0.4

Photograph F

0.8

0.8

0.0

0.4

Photograph E

0.0

0.0

0.4

Photograph H

0.8

0.8

0.0

0.4

Photograph G

er

ts

us

is
nt

er

ie

iv

Sc

ag
er
s
i
sc sta
ho nt
ol hi
ch gh
ild
re
N
n
sc ear
ho by
ol h
ch igh
ild
re
n

an

er
iv
R

Sc

ie

nt

us

is

er

ts

0.0

0.4

Photograph J

0.4
0.2

se
rs

is

er
u

nt
R

iv

ie
Sc

an

ag
e

rs

ts

0.0

No action
Improved landscaping
Wildlife protection
Bank stabilization
Channel cleaning
Engineered structures
Other actions

Total

sc ista
ho nt
ol hi
ch gh
ild
re
N
n
sc ear
ho by
ol h
ch igh
ild
re
n

an

ag

er
s
sc ista
ho nt
ol hi
ch gh
ild
re
N
n
sc ear
ho by
ol h
ch igh
ild
re
n

0.0

0.4

Photograph I

Frequency
Frequency
Frequency

Photograph B

Photograph C

0.0

Frequency

0.0

0.4

Photograph A

0.8

0.8

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Frequency

10

Fig. 7. Categorization of management actions recommended for each riverscape photograph.

reduce the gap between expert and generalist local knowledge

about braided reaches of river? The challenges for sustainable river
management include a social learning based on the polycentric
governance of natural resources, the involvement of all stakeholders in the policy process, and the development of new attitudes
to deal with differences constructively (Mostert et al., 2007).
First, the results of this survey showed that visiting the river
changes the landscape assessment of observers. Survey participants
familiar with the Magra River (local residents and nearby high
school children) provided more consistent opinions regarding
aesthetic value, benecial uses, and management needs than the

distant high school children. Groups familiar with the river chose
the modality no action in terms of need for management more
frequently, and were less tempted to suggest cleaning or engineering measures. This is why any strategy for conserving or
restoring braided segments should be accompanied by the provision of guided tours on-site to explain project objectives. Consulting with the public during the planning stages will be important in
strengthening community support and mitigate potential conicts
(House and Fordham, 1997), certainly, but stakeholder involvement
should be complemented by an environmental educational program on the ow of benecial services coming from braided river

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

habitats. Its success will depend on its duration, teaching practises

employed, and people targeted (Vaughan et al., 2003). The main
objectives for environmental education should be to increase
peoples understanding of how braided rivers function and to
strengthen their emotional afnity with gravel bars.
Furthermore, river manager attitude is also unexpected in the
sense that it differs from those of scientists and other respondents
about vegetation (both in terms of aesthetics and need for
improvement). Additional research is needed to understand the
perception of Italian scientists who do not appreciate braided river
landscape aesthetics (similarly to local residents) but promote their
conservation and restoration. We can hypothesize here that professional training or management policy can inuence their attitude and motivation for improvement. Similarly to what is
observed in other cultural contexts, vegetation and wood in rivers
can thus be interpreted as either a risk factor during oods which
must be controlled or a signicant component of freshwater eco_ et al., 2009).
systems (Wyzga
In contrast to approaches based solely on expert judgement or
developed by small groups of professionals, more and more studies
are collecting perceptions and evaluations of so-called secular
culture. For example, recent studies in public decision-making
considered not only the ecological diversity of a landscape, but
also the aesthetic perceptions of local residents (Cottet, 2010). According to Penning-Rowsell (1981), we need to identify what
people believe are the facets of landscape value, rather than what
the researcher, the historian, the architect and planner think they
believe. This question is particularly challenging in the context of
the implementation of the European Water Framework Directive,
adopted on October 23, 2000, which commits member nations to
restore water and aquatic habitat quality by 2015. With an
emphasis on the benets of public involvement, the Directive requires that river management decisions be made at a social level as
close as possible to the places of use or degradation of the water
and aquatic environments. Accordingly, local communities, water
and river users, non-prot organizations, and government agencies
must all be consulted as part of this stakeholder-based approach.
5. Conclusions
This participatory approach based on a photo-questionnaire can
be used to recognize variations in riverscape perceptions of different
respondents groups and thus to contribute to sustainably managing
rivers and streams. Braided rivers represent high-energy uvial environments where functional processes are positively valued by scientists. The general public however shows a different understanding
of what an ideal river ought to look like. Scientists and managers are
expected to have a deeper technical knowledge of rivers and riverscape. Public support or reaction to restoration projects depends
mainly on the distance between the expert and public perceptions.
Like the in-channel wood, gravel bars are negatively perceived and
often regarded as an alien element in the river. Changing the attitudes
towards bedload seems to be a prerequisite for any plan to reverse the
disappearance of braided reaches in the Magra River. Knowledge is
a really critical factor to consider when reconciling perceptive dissonance and promoting effective preservation and restoration of
unaesthetic riverscapes. Achieving campaigns of environmental education could enhance the recognition of ecological and social benets related to sediment deposits and braided river habitats, and thus
bridge the gap between expert and generalist local knowledge.
Acknowledgements
This study was completed in June 2006, at a summer school
bringing together researchers and doctoral students of the

11

University of Florence (Italy) and the University of Lyon (France).

Data analysis was funded by the ANR GESTRANS (gs1) project
granted to Alain Recking. The authors are particularly grateful to
Dufour S., Rinaldi M., and Simoncini C. who organized the
summer school, the associate eld campaign and additional
exchanges and surveys done thereafter. They also thank Attardo
S., Bellacci L., Comiti F., Coppi L., Doretti G., Gaertner V.,
Grosprtre L., Lamagna S., Lejot J., Luppi L., Mao L., Pecorari E.,
Rollet A.-J., and Vergaro A., who helped distribute the questionnaires and collect the data. R. Jenkinson translated this manuscript
from French to English and A. Stomp edited the translation. The
authors wish to thank the reviewers and Emeline Comby for their
constructive comments and suggestions.
References
Arscott, D., Tockner, K., Nat van der, D., Ward, J.V., 2002. Aquatic habitat dynamics
along a braided alpine river ecosystem (Tagliamento River, Northeast Italy).
Ecosystems 5, 802e814.
Brown, T.C., Daniel, T.C., 1991. Landscape aesthetics of riparian environments: relationships of ow quantity to scenic quality along a wild and scenic river.
Water Resources Research 27, 1787e1795.
Buijs, A.E., 2009. Public support for river restoration. A mixed-method study into
local residents support for and framing of river management and ecological
restoration in the Dutch oodplains. Journal of Environmental Management 90,
2680e2689.
Brunke, M., Gonser, T., 1997. The ecological signicance of exchange processes between rivers and groundwater. Freshwater Biology 37, 1e33.
Buhyoff, G.J., Wellman, J.D., Koch, N.E., Gauthier, L., Hultman, S., 1983. Landscape
preference metrics: an international comparison. Journal of Environmental
Management 16, 181e190.
Bulut, Z., Yilmaz, H., 2009. Determination of waterscape beauties through visual
quality assessment method. Environmental Monitoring and Assessment 154,
459e468.
Church, M., Jones, D., 1992. Channel bars in gravel-bed streams. In: Hey, R.D. (Ed.),
Gravel Bed Streams. Wiley, Chichester, pp. 291e338.
Cossin, M., Pigay, H., 2001. Les photographies prises au sol, une source dinformation pour la gestion des paysages riverains des cours deau. Cahiers de
Gographie du Qubec 45, 37e62.
Cottet, M., 2010. La perception des bras morts uviaux. Le paysage, un mdiateur
pour laction dans le cadre de lingnierie de la restauration. Approche conceptuelle et mthodologique applique aux cas de lAin et du Rhne. PhD thesis,
University of Lyon 3.
Dakin, S., 2003. Theres more to landscape than meets the eye: towards inclusive
landscape assessment in resource and environmental management. The Canadian Geographer 47 (2), 185e200.
Dearden, P., 1981. Public participation and scenic quality analysis. Landscape Planning 8, 3e19.
Ekins, P., Simon, S., Deutsch, L., Folke, C., De Groot, R., 2003. A framework for the
practical application of the concepts of critical natural capital and strong sustainability. Ecological Economics 44, 165e185.
Gardiner, J.L., 1994. Sustainable development for river catchments. Journal of the
Institution of Water and Environmental Management 8 (3), 308e319.
Gray, D., Harding, J.S., 2007. Braided River Ecology: a Literature Review of Physical
Habitats and Aquatic Invertebrate Communities. New Zealand Department of
Conservation, Science for Conservation 279, Wellington.
Hohensinner, S., Habersack, H., Jungwirth, M., Zauner, G., 2004. Reconstruction of
the characteristics of a natural alluvial river-oodplain system and hydromorphological changes following human modications: the Danube River
(1812e1991). River Research and Applications 20 (1), 25e41.
House, M., Fordham, M., 1997. Public perceptions of river corridors and attitudes
towards river works. Landscape Research 22, 25e44.
House, M.A., Sangster, E.K., 1991. Public perception of river-corridor management.
Journal of the Institution of Water and Environmental Management 5, 312e316.
Hull, R.B., Stewart, W.P., 1992. Validity of photo-based scenic beauty judgments.
Journal of Environmental Psychology 12, 101e114.
Ihaka, R., Gentleman, R., 1996. R: a language for data analysis and graphics. Journal
of Computational and Graphical Statistics 5 (3), 299e314.
Junker, B., Buchecker, M., 2008. Aesthetic preferences versus ecological objectives in
river restorations. Landscape and Urban Planning 85, 141e154.
Kenwick, R.A., Shammin, M.R., Sullivan, W.C., 2009. Preferences for riparian buffers.
Landscape and Urban Planning 91, 88e96.
Le Lay, Y.-F., Cottet, M., Pigay, H., Rivire-Honegger, A., 2012. Ground imagery and
environmental perception: using photo-questionnaires to evaluate river management strategies. In: Carbonneau, P., Pigay, H. (Eds.), Fluvial Remote Sensing
for Science and Management. WileyeBlackwell, Chichester, pp. 405e429.
Le Lay, Y.-F., Pigay, H., Gregory, K., Chin, A., Doldec, S., Elosegi, A., Mutz, M.,
_
Wyzga,
B., Zawiejska, J., 2008. Variations in cross-cultural perception of riverscapes in relation to in-channel wood. Transactions of the Institute of British
Geographers 33, 268e287.

12

Y.-F. Le Lay et al. / Journal of Environmental Management 119 (2013) 1e12

Loomis, J., Kent, P., Strange, L., Fausch, K., Covich, A., 2000. Measuring the total
economic value of restoring ecosystem services in an impaired river basin:
results from a contingent valuation survey. Ecological Economics 33, 103e117.
Lothian, A., 1999. Landscape and the philosophy of aesthetics: is landscape quality
inherent in the landscape or in the eye of the beholder? Landscape and Urban
Planning 44, 177e198.
Mostert, E., Pahl-Wostl, C., Rees, Y., Searle, B., Tbara, D., Tippett, J., 2007. Social
learning in European river-basin management: barriers and fostering mechanisms from 10 river basins. Ecology and Society 12 (online) URL: http://www.
ecologyandsociety.org/vol12/iss1/art19/.
Peiry, J.-L., Bravard, J.-P., 1989. Evolution naturelle dun remplissage sdimentaire
intramontagnard et impacts des amnagements contemporains. Houille Blanche 3e4, 221e225.
Penning-Rowsell, E.C., 1981. Fluctuating fortunes in gauging landscape value.
Progress in Human Geography 5, 25e41.
Petrosillo, I., Costanza, R., Aretano, R., Zaccarelli, N., Zurlini, G., 2013. The use of
subjective indicators to assess how natural and social capital support residents
quality of life in a small volcanic island. Ecological Indicators 24, 609e620.
Pger, Y., Rackham, A., Larned, S., 2010. The aesthetic value of river ows: an
assessment of ow preferences for large and small rivers. Landscape and Urban
Planning 95, 68e78.
Pigay, H., Alber, A., Slater, L., Bourdin, L., 2009. Census and typology of the braided
rivers in the French Alps. Aquatic Sciences 71 (3), 371e388.
Porteous, J.D., 1996. Environmental Aesthetics: Ideas, Politics and Planning. Routledge, London.
Raymond, M.C., Bryan, B.A., MacDonald, D.H., Cast, A., Strathearn, S., Grandgirard, A.,
Kalivas, T., 2009. Mapping community values for natural capital and ecosystem
services. Ecological Economics 68, 1301e1315.
Rinaldi, M., Simoncini, C., 2006. Studio geomorfologico del F. Magra e del F. Vara
nalizzato alla gestione dei sedimenti e della fascia di mobilit. In: Autorit di
Bacino Interregionale del ume Magra (Ed.), Nuovi approcci per la comprensione
dei processi uviali e la gestione dei sedimenti e Applicazioni nel Bacino del F.
Magra. Autorit di Bacino Interregionale del ume Magra, Sarzana, pp. 93e110.
Rinaldi, M., Simoncini, C., Pigay, H., 2009. Scientic design strategy for promoting
sustainable sediment management: the case of the Magra River (CentralNorthern Italy). River Research and Applications 25, 607e625.
Rinaldi, M., Simoncini, C., Sogni, D., 2005a. Variazioni morfologiche recenti di due
alvei ghiaiosi appenninici: il Fiume Trebbia ed il Fiume Vara. Supplementi di
Geograa Fisica e Dinamica Quaternaria VII, 313e319.
Rinaldi, M., Wyzga, B., Surian, N., 2005b. Sediment mining in alluvial channels:
physical effects and management perspectives. River Research and Applications
21, 805e828.

Ryan, R.L., 1998. Local perceptions and values for Midwestern river corridor.
Landscape and Urban Planning 42, 225e237.
Sambrook Smith, G.H. (Ed.), 2006. Braided Rivers: Process, Deposits, Ecology and
Management. IAS Special Publication 36, Blackwell Science, Oxford.
Schroeder, H.W., 1984. Environmental perception rating scales. A case for simple
methods of analysis. Environment and Behavior 16, 573e598.
Shuttleworth, S., 1980. The use of photographs as an environment presentation
medium in landscape studies. Journal of Environmental Management 23, 285e
305.
Simoncini, C., 2008. Studio delle problematiche legate ai sedimenti uviali ed
applicazione di una metodologia nalizzata alla gestione a scala di bacino
idrograco. PhD thesis, University of Florence.
Stamps III, A.E., 1992. Bootstrap investigation of respondent sample size for environmental preference. Perceptual and Motor Skills 75, 220e222.
Stamps III, A.E., 1990. Use of photographs to simulate environments. A meta-analysis. Perceptual and Motor Skills 71, 907e913.
Surian, N., 1999. Channel changes due to river regulation: the case of the Piave
River, Italy. Earth Surface Processes and Landforms 24, 1125e1151.
Surian, N., Rinaldi, M., 2003. Morphological response to river engineering and
management in alluvial channels in Italy. Geomorphology 50, 307e326.
Tockner, K., Paetzold, A., Karaus, U., Claret, C., Zettel, J., 2006. Ecology of braided
rivers. In: Sambrook Smith, G.H. (Ed.), Braided Rivers: Process, Deposits, Ecology and Management. Wiley and Blackwell, Publication n 36 of the International Association of Sedimentologists (IAS), Oxford, pp. 339e359.
Vaughan, C., Gack, J., Solorazano, H., Ray, R., 2003. The effect of environmental
education on schoolchildren, their parents, and community members: a study
of intergenerational and intercommunity learning. Journal of Environmental
Education 34, 12e21.
Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments.
Journal of Geology 30, 377e392.
Whited, D.C., Lorang, M.S., Harner, M.J., Hauer, F.R., Kimball, J.S., Stanford, J.A., 2007.
Climate, hydrologic disturbance, and succession: drivers of oodplain pattern.
Ecology 88, 940e953.
Wilson, M.A., Carpenter, S.R., 1999. Economic valuation of freshwater ecosystem
services in the United States: 1971e1997. Ecological Applications 9, 772e783.
Wohl, E., Merritts, D.J., 2007. What is a natural river? Geography Compass 1, 871e
900.
_
Wyzga,
B., Zawiejska, J., Le Lay, Y.-F., 2009. Inuence of academic education on the
perception of wood in watercourses. Journal of Environmental Management 90,
587e603.
Zube, E.H., Sell, J., Taylor, J., 1982. Landscape perception: research, application and
theory. Landscape Planning 9, 1e32.