Environmental Management (2014) 54:194–204

DOI 10.1007/s00267-014-0290-2

Mediterranean Coastal Sand Dune Vegetation: Influence

of Natural and Anthropogenic Factors
Daniela Ciccarelli

Received: 16 September 2013 / Accepted: 26 April 2014 / Published online: 15 May 2014
Ó Springer Science+Business Media New York 2014

Abstract The aim of the present work was to assess the could be to protect the foredunes from erosion and limit
conservation status of coastal dune systems in Tuscany trampling through the installation of footbridges or the use
(Italy). Emphasis was given to the presence and abundance of appropriate fences.
of plant communities identified as habitat in accordance
with the Directive 92/43/EEC. Twenty transects perpen- Keywords Canonical Correspondence Analysis  Coastal
dicular to the shoreline were randomly positioned on the dune vegetation  Disturbance  Management  Natura
whole coastal area (30 km in length) in order to sample the 2000  Principal Coordinate Analysis
full spectrum of plant communities. Vegetation zonation
and relationships with the most frequent disturbance fac-
tors in the study area—beach cleaning, coastline erosion, Introduction
presence of paths and roads, bathing settlements and
trampling—were investigated through principal coordinate Coastal dune habitats are undeniably important ecosystems
analysis and canonical correspondence analysis. Natural in Europe, as they contain a high ecological diversity in
factors, such as distance from the sea and total length, were terms of environmental heterogeneity and variability of
also considered. Differences in the conservation status of species composition (Van der Maarel 2003; Martı́nez et al.
the sites were found, ranging from the total disappearance 2004). They are characterized by a complex coast-to-inland
of the foredune habitats to the presence of the complete environmental gradient which determines the coexistence
psammophilous (sand-loving) plant communities. Erosion, of different plant communities in a relatively small area
trampling, and paths were found to be closely correlated (Ranwell 1972; Doing 1985; Psuty 2004; Wiedemann and
with degradation and habitat loss. Furthermore, the overall Pickart 2004; Frederiksen et al. 2006). However, many
plant species diversity of dunes was measured with factors can disturb coastal ecosystems worldwide, such as
NHDune, a modified version of the Shannon index; while pollution, coastal erosion, effects of global warming,
the incidence of invasive taxa was calculated using N, a farming practices, urban development, and pressure from
naturalness index. However, these diversity indices proved tourism (Cori 1999; Brown and McLachlan 2002; Van der
to be a weaker bioindicator of ecosystem integrity than Maarel 2003; Coombes et al. 2008; Carboni et al. 2009;
habitat composition along transects. A possible strategy for Gornish and Miller 2010; Miller et al. 2010; Ciccarelli
the conservation and management of these coastal areas et al. 2012).
Although the definition of conservation status has been
debated, it is generally recognized that direct or indirect
Electronic supplementary material The online version of this
article (doi:10.1007/s00267-014-0290-2) contains supplementary measures of biodiversity can be used to assess it (Margules
material, which is available to authorized users. and Pressey 2000; Cabeza and Moilanen 2001; Yoccoz
et al. 2001). Regarding coastal sand dunes, many methods
D. Ciccarelli (&)
for conservation status assessment have been proposed
Department of Biology, University of Pisa, Via L. Ghini 13,
56126 Pisa, Italy focusing on: single species as bioindicators or parameters,
e-mail: [email protected] such as species composition, richness, diversity and cover

123
Environmental Management (2014) 54:194–204 195

(Van der Maarel and Van der Maarel-Versluys 1996; presence of paths and settlements, and trampling—on the
Kutiel 2001; Grunewald and Schubert 2007); plant com- vegetation along the coast-to-inland gradient in sand dune
munities as bioindicators of the ecosystem integrity systems and, in particular, to correlate EU habitat of
(Provoost et al. 2004; Acosta et al. 2007; Lomba et al. interest (in accordance with Council of European Com-
2008; De Luca et al. 2011; Attorre et al. 2013); landscape munities 1992) with disturbance factors typical of coastal
cover types and related structural indices (Carboni et al. ecosystems. The second objective was to assess the con-
2009, Drius et al. 2013, Malavasi et al. 2013); integrated servation status of coastal dune systems belonging to two
dune vulnerability index based on geomorphological con- regional parks along the Tuscan littoral (Italy), which are
dition, marine influence, aeolian factor, vegetation char- characterized by the presence of the most well-preserved
acteristics, and human effects (Garcı́a-Mora et al. 2001; dune systems of Tuscany. In particular, vegetation surveys
Martı́nez et al. 2006; Muñoz Vallés et al. 2011). In par- allowed identification of the typical psammophilous plant
ticular, to better understand the effect of human distur- communities, their sequence along transects, and calcula-
bance on plant communities specifically of dune tion of suitable diversity indices that are useful to assess the
ecosystems, several studies have highlighted that moderate integrity of coastal dune ecosystems. Because the Medi-
to high human trampling intensity can decrease plant terranean plant communities of the sandy coastal systems
diversity, cover, and productivity especially regarding rare are particularly endangered, the need to assess the con-
and threatened coastal plants (Andersen 1995; Kutiel et al. servation status of coastal dunes has become urgent in
1999; Lemauviel and Rozé 2003; Kerbiriou et al. 2008; order to promote appropriate management strategies to
Santoro et al. 2012; Farris et al. 2013; Fenu et al. 2013b). preserve these environments (Rivas-Martinez 2005; Men-
The Mediterranean basin has been intensively affected by doza-González et al. 2012).
human populations for thousands of years especially along
the coasts (Lavorel et al. 1998; Cori 1999; Falcucci et al.
2007). The problem of the human alteration of coastal Materials and Methods
habitats is very widespread along the Mediterranean area as
a result of urbanization, industrialization, and tourism Study Area
activities (EEA 1999). During the last century, in fact, a loss
of about 70 % of dune systems for European coasts has been The present study was conducted in the coastal sand dunes
calculated, and in Italy the process of dune degradation of two protected areas along the Tuscan littoral (Italy):
seems to be even more intense than the European average: Migliarino-San Rossore-Massaciuccoli Regional Park and
up to 80 %, from 45,000 ha in 1900 to a remaining 9,000 ha Maremma Regional Park (Fig. 1).
at the end of the twentieth century (Feola et al. 2011). In The Migliarino-San Rossore-Massaciuccoli Regional
addition, the Mediterranean Sea has a unique character that Park (San Rossore) covers an area of 142 km2 and is
results from its nature as a semi-enclosed basin surrounded located near Pisa, in the North of Tuscany. The Park hosts
by a complex orography, which strongly affects and often 40 km2 of coastal forest, sandy beach, inland marshes, and
controls the local climate. It is characterized by high water farming areas. This area is characterized by a Mediterra-
temperature and salinity, and very limited tides, due to the nean sub-humid climate (C2), with a mean annual tem-
scarce exchange with the low-salinity water from the perature [15 °C and a mean annual rainfall of 800 mm
Atlantic Ocean and especially to the high evaporation rates (Rapetti 2003).
(Weyl 1970; King 1975). Moreover, Mediterranean clima- The Maremma Regional Park (Maremma) is located on
tology is peculiar also because of confined air circulation the coast near Grosseto, in the southern part of Tuscany.
and strong seasonal variability (D’Ortenzio et al. 2005). The Park is about 90 km2 and offers a diverse array of
Recently, research has made evident not only floristic, landscape types: coastal dunes, inland salt marshes, rocky
structural, and ecological differences between the Atlantic vegetated foothills, coastal pinewood, and farming areas.
and the Mediterranean plant communities of coastal dunes This area is characterized by a C1 type of climate, Medi-
(Feola et al. 2011), but also the minor importance of wind terranean sub-arid, with an average annual temperature of
parameters as factors controlling the vegetation pattern 15.6 °C and an average annual rainfall of 618 mm (Pinna
along the Mediterranean dune systems with respect to the 1985).
oceanic ecosystems (Fenu et al. 2013a). The coastal dune systems of both Migliarino-San Ros-
In this paper a systematic sampling design using belt sore-Massaciuccoli Regional Park and Maremma Regional
transects was followed in order to sample the full spectrum Park belong to the Natura 2000 network and include the
of coastal sand dune plant communities. The first aim of following Sites of Community Importance (SCIs):
the present work was to analyze the effects of natural and ‘‘Coastal sand dunes of Torre del Lago’’ and ‘‘Coastal sand
anthropogenic factors—such as beach cleaning, erosion, dunes of Uccellina,’’ respectively.

123
Environmental Management (2014) 54:194–204 195

(Van der Maarel and Van der Maarel-Versluys 1996; presence of paths and settlements, and trampling—on the
Kutiel 2001; Grunewald and Schubert 2007); plant com- vegetation along the coast-to-inland gradient in sand dune
munities as bioindicators of the ecosystem integrity systems and, in particular, to correlate EU habitat of
(Provoost et al. 2004; Acosta et al. 2007; Lomba et al. interest (in accordance with Council of European Com-
2008; De Luca et al. 2011; Attorre et al. 2013); landscape munities 1992) with disturbance factors typical of coastal
cover types and related structural indices (Carboni et al. ecosystems. The second objective was to assess the con-
2009, Drius et al. 2013, Malavasi et al. 2013); integrated servation status of coastal dune systems belonging to two
dune vulnerability index based on geomorphological con- regional parks along the Tuscan littoral (Italy), which are
dition, marine influence, aeolian factor, vegetation char- characterized by the presence of the most well-preserved
acteristics, and human effects (Garcı́a-Mora et al. 2001; dune systems of Tuscany. In particular, vegetation surveys
Martı́nez et al. 2006; Muñoz Vallés et al. 2011). In par- allowed identification of the typical psammophilous plant
ticular, to better understand the effect of human distur- communities, their sequence along transects, and calcula-
bance on plant communities specifically of dune tion of suitable diversity indices that are useful to assess the
ecosystems, several studies have highlighted that moderate integrity of coastal dune ecosystems. Because the Medi-
to high human trampling intensity can decrease plant terranean plant communities of the sandy coastal systems
diversity, cover, and productivity especially regarding rare are particularly endangered, the need to assess the con-
and threatened coastal plants (Andersen 1995; Kutiel et al. servation status of coastal dunes has become urgent in
1999; Lemauviel and Rozé 2003; Kerbiriou et al. 2008; order to promote appropriate management strategies to
Santoro et al. 2012; Farris et al. 2013; Fenu et al. 2013b). preserve these environments (Rivas-Martinez 2005; Men-
The Mediterranean basin has been intensively affected by doza-González et al. 2012).
human populations for thousands of years especially along
the coasts (Lavorel et al. 1998; Cori 1999; Falcucci et al.
2007). The problem of the human alteration of coastal Materials and Methods
habitats is very widespread along the Mediterranean area as
a result of urbanization, industrialization, and tourism Study Area
activities (EEA 1999). During the last century, in fact, a loss
of about 70 % of dune systems for European coasts has been The present study was conducted in the coastal sand dunes
calculated, and in Italy the process of dune degradation of two protected areas along the Tuscan littoral (Italy):
seems to be even more intense than the European average: Migliarino-San Rossore-Massaciuccoli Regional Park and
up to 80 %, from 45,000 ha in 1900 to a remaining 9,000 ha Maremma Regional Park (Fig. 1).
at the end of the twentieth century (Feola et al. 2011). In The Migliarino-San Rossore-Massaciuccoli Regional
addition, the Mediterranean Sea has a unique character that Park (San Rossore) covers an area of 142 km2 and is
results from its nature as a semi-enclosed basin surrounded located near Pisa, in the North of Tuscany. The Park hosts
by a complex orography, which strongly affects and often 40 km2 of coastal forest, sandy beach, inland marshes, and
controls the local climate. It is characterized by high water farming areas. This area is characterized by a Mediterra-
temperature and salinity, and very limited tides, due to the nean sub-humid climate (C2), with a mean annual tem-
scarce exchange with the low-salinity water from the perature [15 °C and a mean annual rainfall of 800 mm
Atlantic Ocean and especially to the high evaporation rates (Rapetti 2003).
(Weyl 1970; King 1975). Moreover, Mediterranean clima- The Maremma Regional Park (Maremma) is located on
tology is peculiar also because of confined air circulation the coast near Grosseto, in the southern part of Tuscany.
and strong seasonal variability (D’Ortenzio et al. 2005). The Park is about 90 km2 and offers a diverse array of
Recently, research has made evident not only floristic, landscape types: coastal dunes, inland salt marshes, rocky
structural, and ecological differences between the Atlantic vegetated foothills, coastal pinewood, and farming areas.
and the Mediterranean plant communities of coastal dunes This area is characterized by a C1 type of climate, Medi-
(Feola et al. 2011), but also the minor importance of wind terranean sub-arid, with an average annual temperature of
parameters as factors controlling the vegetation pattern 15.6 °C and an average annual rainfall of 618 mm (Pinna
along the Mediterranean dune systems with respect to the 1985).
oceanic ecosystems (Fenu et al. 2013a). The coastal dune systems of both Migliarino-San Ros-
In this paper a systematic sampling design using belt sore-Massaciuccoli Regional Park and Maremma Regional
transects was followed in order to sample the full spectrum Park belong to the Natura 2000 network and include the
of coastal sand dune plant communities. The first aim of following Sites of Community Importance (SCIs):
the present work was to analyze the effects of natural and ‘‘Coastal sand dunes of Torre del Lago’’ and ‘‘Coastal sand
anthropogenic factors—such as beach cleaning, erosion, dunes of Uccellina,’’ respectively.

123
196 Environmental Management (2014) 54:194–204

Fig. 1 a Location of the

Migliarino-San Rossore-
Massaciuccoli (MSRM)
Regional Park and the
Maremma Regional Park in the
northern and southern parts of
Tuscany (Italy), respectively.
On the right a picture of Italy
with Tuscany. b Distribution of
transects along the coast of
MSRM Regional Park.
c Distribution of transects along
the coast of Maremma Regional
Park

Vegetation Sampling respectively) was divided into sections of 1 km and one

random transect was selected within each section (Table
Coastal sand dune systems generally consist of distinct S1). Several coastal traits, which were located in close
plant communities that occur along a coast-to-inland eco- proximity to the mouth of the River Arno and the River
logical gradient. Most of these plant communities corre- Ombrone (in San Rossore Park and in Maremma Park,
spond to habitat types included in the 92/43/EEC Directive respectively), were excluded from this study because there
(Habitat Directive), which specifies a list of habitats and were practically no longer any sand dune systems, as a
species of conservation interest in Europe (Council of result of the strong coastal erosion. Twenty transects
European Communities 1992). This paper focused on the (Fig. 1) orthogonal to the seashore were located in the
plant communities that go from annual vegetation of drift study area (13 in San Rossore Park and seven in Maremma
lines (EC 1210), through the embryonic shifting dunes (EC Park), in order to investigate dune zonation starting from
2110) and shifting dunes along the shoreline with Ammo- the annual vegetation of drift lines to the herbaceous veg-
phila arenaria (EC 2120), as well as the Crucianellion etation of fixed dunes. The length of each transect varied
maritimae fixed beach dunes (EC 2210), and the Malcol- depending on dune morphology and width. The percent
mietalia dune grasslands (EC 2230). Among coastal dune cover of each vascular plant present was visually estimated
habitats, these vegetation types are highly exposed to in contiguous 1 m 9 1 m plots. This size was considered
physical limiting factors and to anthropogenic pressure. sufficient for the herbaceous vegetation analyzed in this
In each protected area, the whole coastal system (20 km study, mainly to avoid overlapping between samples from
and 10 km in length for San Rossore and Maremma park, different plant communities due to compressed zonation

123
Environmental Management (2014) 54:194–204 197

that frequently occurs in Mediterranean coastal dune eco- information criterion (AIC; Akaike 1974). The lower the
systems (especially in disturbed sites). Field work took AIC value the more parsimonious was the model.
place between May 2010 and August 2011. The taxonomic Variance partitioning (Borcard et al. 1992) was per-
nomenclature follows the Checklist of the Italian vascular formed to determine the relative influence of natural and
flora (Conti et al. 2005, 2007) for native species; and Ar- anthropogenic factors using the method and software
rigoni and Viegi (2011) for alien plants. described in Peres-Neto et al. (2006). The software allows
In accordance with Annex E of 97/266/EEC Directive partitioning of the variance in the habitat data coverage
(Council of European Communities 1997), for each transect into components accounted for by natural and anthropo-
the following five disturbance factors affecting the conser- genic factors and their combined effects. The results were
vation status of the plant communities were recorded: represented as a Venn diagram (Økland 2003). PCO, CCA,
cleaning of the beach (cle), erosion of the coastline (ero), and variance partitioning analyses were performed using
presence of paths (pat), bathing settlements (set), and vegan package within the R statistical language software
trampling (tra). In particular, the environmental parameters (Oksanen et al. 2011).
were recorded using the following categories: (0) absent, (1) Spatial autocorrelation of the habitat composition in
low, (2) moderate, (3) high. Moreover, distance from the sea neighboring transects was evaluated following the meth-
(dis) and total length (len) of each transect (expressed in odology proposed by Legendre and Fortin (1989), and
meters) were analyzed as natural features. Each factor was using ncf package within the R statistical language soft-
measured or estimated in the field, except for erosion that ware (Bjornstad 2013). To estimate if the spatial depen-
was deduced from literature (Gruppo Nazionale per la ric- dence between transects was statistically significative,
erca sull’ambiente costiero 2006; Anfuso et al. 2011). Moran’s I correlograms were computed. A matrix of geo-
graphical coordinates for each transect was correlated to a
Data Analysis matrix z of percentage of habitat coverage found in each
transect. Each correlogram was tested for global signifi-
All plots were classified according to plant community cance at the 5 % level using Bonferroni procedures to
composition by hierarchical agglomerative cluster analysis correct for the dependence among the autocorrelation
with the Bray–Curtis distance as a resemblance measure coefficients calculated for each distance class.
and average linkage as a clustering algorithm (Podani All the statistical analyses were performed using the R
2007). Habitats were identified according to the Italian 2.14.0 language software (R Development Core Team, 2011).
Interpretation Manual of the 92/43/EEC Directive Habitat
(Biondi et al. 2009) in order to obtain coverage percentage Diversity Indices
of each habitat in each transect. Several plots were not
classifiable within EC habitats, because the typical diag- In order to measure the overall plant species diversity of
nostic taxa were missing or there were open sands, paths, each transect a modified version of the Shannon index (H),
and/or bathing settlements. They were generically defined named Hdune (Grunewald and Schubert 2007), was used:
as disturbed plots. X
Principal Coordinate Analysis (PCO) was performed to HDune ¼  pi x lnðpi Þ;
investigate variation in species composition among the where pi = % cover of the ith species.
different habitats. The dissimilarity matrix calculated for It has been demonstrated that Hdune is more useful than
PCO was determined using the Bray–Curtis measure on the the Shannon diversity index (Shannon and Weaver 1949) in
square-root transformation of species cover abundance. extreme habitats such as the coastal dunes (De Luca et al.
PCO is an ordination method that takes a symmetric matrix 2011; Attorre et al. 2013), where natural stressful condi-
of plot-to-plot dissimilarities and provides a set of tions determine the presence of a few species with high
Euclidean coordinates for each plot so that the Euclidean dominance (Martı́nez et al. 2004). Hdune, in fact, uses the
distance between two plots in the full-dimensional princi- abundance of species (as cover percentage) in relation to a
pal coordinate space is equal to the original dissimilarity constant sampling area, and unlike H is able to detect
value between both plots (Gower 1966; Legendre and changes in both species diversity and total cover.
Anderson 1999). In the present paper the same methods used by Attorre
Relationships between habitat data coverage and envi- et al. (2013) were followed. First, Hdune was normalized by
ronmental factors recorded in transects were analyzed the logarithm of the species number, because it is directly
using Canonical Correspondence Analysis (CCA; ter Braak dependent on it, as:
1986). The most parsimonious model was fitted using a
1 X
stepwise algorithm, through forward and backward selec- NHDune ¼  pi x ln ðpi Þ;
tion of explanatory variables, based on the akaike lnðkÞ

123
198 Environmental Management (2014) 54:194–204

where k indicates the number of sampled species. shifting dunes (EC 2110) were preferentially located at the
Second, in order to calculate NHDune for each transect, higher end of PCO1, while plots from C. maritimae fixed
the estimated cover of each plant species was normalized dunes (EC 2210) were distributed at the lower PCO1 val-
with respect to the largest sampling area that, in this case, ues. Along PCO2, plots from shifting dunes with A. are-
was 90 m2, since the maximum transect length was 90 m naria (EC 2120) and from Malcolmietalia grasslands (EC
formed by 90 plots of 1 m2 each (see Table S1). In this 2230) were negatively and positively correlated with the
way, transects along well-developed dune systems, with a axis, respectively. Disturbed plots seemed to show a neg-
complete sequence of diverse habitats, are expected to be ative correlation with PCO1. Only four taxa showed a high
characterized by higher NHDune values than transects with correlation with the first two PCO axes (Spearman’s
reduced length or with a low plant species cover due to coefficient [ 0.6): A. arenaria, characteristic species of
disturbance factors. mobile dunes (EC 2120), was related to the lower values of
Moreover, a natural diversity index (N) was also cal- the first PCO axis; Elymus farctus, diagnostic species of
culated, taking into account the percentage cover of alien embryo dunes (EC 2110), was linked to the higher values
species along each transect, in accordance with the fol- of PCO1; while Helichrysum stoechas and Vulpia fascic-
lowing formula (Grunewald and Schubert 2007): ulata, characteristic species of C. maritimae fixed dunes
(EC 2210) and Malcolmietalia grasslands (EC 2230)
NHDune ðwithout alien speciesÞ respectively, were positively correlated with PCO2
N¼ :
NHDune (Fig. 2).
The following taxa, found in transects, were considered The CCA analysis highlighted that erosion, trampling,
aliens for Tuscany: Arundo donax, Cuscuta cesatiana, El- and paths were the environmental factors that significantly
eagnus angustifolia, Erigeron canadensis, Oenothera bi- affected dune habitats accounting for 41.6 % of explained
ennis s.l., Xanthium orientale subsp. italicum (Arrigoni and variance (Table 3; Fig. 3). Even if there was a low spread
Viegi 2011). Spearman’s correlation was performed to of the different habitat types around the origin, erosion
evaluate the effects of single disturbance factors on NHDune seemed to affect especially C. maritimae fixed dunes (EC
and N indices. 2210), while trampling and paths were most related to
embryonic shifting dunes (EC 2110). Based on the AIC
value, the best model included only the erosion parameter
Results (Table 3). The other ecological factors considered in the
present study (beach cleaning, transect length, distance
A total number of 64 vascular plant taxa was recorded in from the sea, and bathing settlements) did not seem to
the 20 transects along the Tuscan coast representing five affect vegetation composition.
plant community types associated to foredunes and fixed Variance partitioning (Fig. 4) confirmed that the envi-
dunes (Tables 1, S2). In the study area all regional typical ronmental variables considered explained a low percentage
dune habitats were found. In particular, the entire habitat (24 %) of the total variability. Natural variables explained
series was present only in transect 15 (5 % of total), while 9 % of total variability, while the anthropogenic factors
the combination of three-four plant communities was found only 1 %. The combined effects of the two groups of
in 14 transects (70 %), and only five transects (25 %) had variables represented 14 % of the total variability.
two plant communities (Table 2). Vegetation of the drift- All the Moran’s I spatial correlograms regarding the five
line (EC 1210) was present in only eight transects (40 %), different habitats in neighboring transects were not sig-
while the embryonic shifting dunes (EC 2110) and shifting nificant (P value [ 0.05 after Bonferroni corrected test),
dunes with A. arenaria (EC 2120) were present in 15 which did not indicate any spatial autocorrelations between
(75 %) and 13 transects (65 %), respectively (Table 2). C. the data (Table 4).
maritimae fixed dunes (EC 2210) and Malcolmietalia Regarding diversity indices (Table 5), transect 17
grasslands (EC 2230) were found in 14 (70 %) and 12 showed the lowest NHDune value (NHDune = 0.47), while
transects (60 %), respectively (Table 2). Lastly, 16 tran- transect 12 had the highest value (NHDune = 0.84). Tran-
sects (80 %) were characterized by the presence of dis- sect 15, which was the only one with the whole habitat
turbed vegetation (Table 2). series, showed a very high NHDune value (NHDune = 0.83).
The PCO, performed on the plot-to-plot dissimilarity Moreover, the naturalness index (N) was equal to 1 in
matrix, resulted in a good distinction of the different hab- seven transects (35 % of total) and very high in the others
itats in the multivariate space, although there was a certain (Table 5). Spearman’s correlation performed on single
degree of overlap between several groups (Fig. 2). Plots disturbance factor and diversity indices of each transect did
from vegetation of the driftline (EC 1210) and embryonic not evidence any significant statistical effects (P [ 0.05).

123
Environmental Management (2014) 54:194–204 199

Table 1 EU interest habitat types and plant communities found in the study areas along the sea-inland gradient
Upper beach Embryonic dune Mobile dune Transition dune Backdune
and/or backdune

EU Habitat 1210—annual vegetation of 2110—embryonic shifting 2120—shifting 2210— 2230—Malcolmietalia

drift lines dunes dunes along Crucianellion dune grasslands
the shoreline maritimae fixed
with beach dunes
Ammophila
arenaria
(white dunes)
Plant Formations of annuals or Formations of the coast Mobile dunes Perennial Formations of small
community representatives of annuals representing the first forming the community of annuals and ephemeral
description and perennials, occupying stages of dune seaward the inland side of spring bloom plants of
accumulations of drift construction, constituted cordon or mobile dunes, deep sands in dry
material and gravel rich in by ripples or raised sand cordons of dominated by interdunal depressions
nitrogenous organic matter surfaces of the upper dune systems chamaephytic of the coasts
beach or by a seaward of the coasts species
fringe at the foot of the tall
dunes
Main Cakile maritima, Salsola Elymus farctus subsp. Ammophila Crucianella Malcolmia ramosissima,
diagnostic kali farctus, Sporobolus arenaria maritima, Silene canescens,
speciesa virginicus Pycnocomon Vulpia fasciculata
rutifolium, Seseli
tortuosum
a
According to the Italian Interpretation Manual of EU Habitats (Biondi et al. 2009)

Table 2 Percentage of Transect Upper beach Embryonic Mobile dune Transition dune Backdune Disturbed
coverage of habitats found in (EC 1210) dune (EC 2110) (EC 2120) (EC 2210) (EC 2230) habitat
each transect
1 0.0 58.2 0.0 7.6 31.6 2.5
2 0.0 47.4 7.9 10.5 32.9 1.3
3 8.6 45.7 0.0 15.7 27.1 2.9
4 5.4 59.5 0.0 0.0 29.7 5.4
5 0.0 85.2 0.0 7.4 7.4 0.0
6 0.0 6.7 21.1 70.0 2.2 0.0
7 0.0 0.0 53.8 36.9 1.5 7.7
8 33.3 0.0 0.0 20.8 29.2 16.7
9 0.0 36.4 0.0 30.3 0.0 33.3
10 0.0 0.0 72.4 20.7 0.0 6.9
11 0.0 21.7 56.5 0.0 0.0 21.7
12 0.0 11.1 88.9 0.0 0.0 0.0
13 2.1 68.1 29.8 0.0 0.0 0.0
14 0.0 78.0 0.0 17.1 4.9 0.0
15 18.5 22.2 22.2 7.4 11.1 18.5
16 59.4 7.2 11.6 0.0 7.2 14.5
17 17.1 0.0 42.9 34.3 0.0 5.7
18 6.3 0.0 60.4 33.3 0.0 0.0
19 0.0 57.1 12.7 27.0 3.2 0.0
For definitions of habitat codes 20 0.0 58.5 19.5 0.0 0.0 22.0
see Table 1

Discussion dunes (e.g., Stallins 2002; Acosta et al. 2007; Forey et al.
2008; Miller et al. 2010; Fenu et al. 2012). The analysis of
The general spatial pattern of the vegetation found in this coverage habitat data (Table 2) indicated that transects
study was consistent with previous studies on coastal sand exhibit varying pattern of the typical sequence of coastal

123
200 Environmental Management (2014) 54:194–204

Fig. 2 PCO based on the dissimilarity (measured by the Bray–Curtis

dissimilarity metric) occurring in species composition among the 922
sampled plots. Species with a Spearman correlation coefficient [0.6
with the first two PCO axes are shown. Plant species abbreviations:
Ammare, Ammophila arenaria; Elyfar, Elymus farctus; Helsto,
Fig. 3 CCA plot of the data matrix of habitat coverage and
Helichrysum stoechas; Vulfas, Vulpia fasciculata. Habitat codes:
environmental factors. Numbers indicate transects and arrows repre-
1210 annual vegetation of drift lines (upper beach), 2110 embryonic
sent the environmental factors. F1 1210 (annual vegetation of drift
shifting dunes, 2120 shifting dunes along the shoreline with
lines), F2 2110 (embryonic shifting dunes), F3 2120 (shifting dunes
Ammophila arenaria (white dunes), 2210 Crucianellion maritimae
along the shoreline with Ammophila arenaria), F4 2210 (Crucianel-
fixed beach dunes, 2230 Malcolmietalia dune grasslands. The first
lion maritimae fixed beach dunes), F5 2230 (Malcolmietalia dune
axis (PCO1) explained 19.9 % of total variation, while PCO2 14.6 %
grasslands). Cle cleaning of the beach, Dis distance from the sea, Ero
erosion, Len length of the transect, Pat paths, Set bathing settlements,
Tra trampling. Eigenvalues for constrained axes: CCA1 0.27 and
Table 3 Results of the CCA analysis (total inertia: 1.2783) CCA2 0.15
Explanatory variables considered in the model AIC P value

?Erosion 97.274 0.005**

?Trampling 98.133 0.020*
?Paths 98.576 0.050*
?Distance from the sea 99.013 0.100
?Cleaning 99.594 0.310
?Length of transect 99.852 0.350
?Settlements 99.861 0.340
AIC value of akaike information criterion
*P \ 0.05; **P \ 0.01

dune plant communities, ranging from the disappearance of

the foredune habitats (EC 1210, 2110 and 2120) to the
presence of the complete sequence. Vegetation of the
driftline (EC 1210) disappeared in those transects where
erosion was more consistent, while embryonic shifting Fig. 4 A Venn diagram showing the variance partitioning of
dunes with E. farctus (EC 2110) were well represented in conditional and shared effects of the groups of natural and anthro-
the area. However, embryonic dune habitat showed a pogenic variables as fractions of the total inertia
decrease of the coverage in transects belonging to coastal
tracts affected by erosion. These results are in accordance
with data present in literature (Lomba et al. 2008; De Luca Table 4 Results of Moran’s I test to check for spatial autocorrelation
et al. 2011; Attorre et al. 2013) that correlated absence of regarding the five different habitats found in the study area
embryonic foredunes with erosional phenomena. Even
EU Habitat 1210 2110 2120 2210 2230
though vegetation of the driftline and embryonic shifting
Moran’s I 0.296 0.250 0.253 0.253 0.257
dunes are poor of taxa with low coverage, they represent an
important element that must be considered in conservation P values are reported. For each test 999 permutations were performed.
programs (Acosta et al. 2009). For definitions of habitat codes see Table 1

123
Environmental Management (2014) 54:194–204 201

Table 5 Diversity indices Transect NHdune N fixed dunes, where these habitats tend to have a mosaic
values for each transect pattern of distribution.
1 0.71 0.99 Both through the CCA and variance partitioning, I found
2 0.80 0.98 that natural variables explained more of the variation
3 0.72 0.96 observed compared to the anthropogenic factors. Similarly
4 0.67 0.97 to other studies on Italian coasts (De Luca et al. 2011;
5 0.59 0.93 Attorre et al. 2013), the CCA analysis revealed that the
6 0.52 1.00 most important variables in the studied habitat-environ-
7 0.58 1.00 ment relations were erosion, trampling, and paths (Fig. 3).
8 0.77 0.93 Surprisingly, the other anthropogenic factors considered in
9 0.63 1.00 this study, beach cleaning and bathing settlements, did not
10 0.74 1.00 seem necessarily to coincide with habitat decay. The
11 0.60 1.00 present results might depend on a relatively good man-
12 0.84 1.00 agement of natural resources by the San Rossore and
13 0.64 0.99 Maremma parks that host the most well-preserved dune
14 0.80 0.99 systems of Tuscany. In any case this study showed that
15 0.83 0.98 coastal dynamic processes of the Tuscan coast may be
16 0.78 0.99 stronger and have more important effects on vegetation
NHDune is a modified version of composition and distribution. Beach erosion causes insta-
17 0.47 1.00
the Shannon index; while N is a
naturalness index that calculates 18 0.51 0.98 bility and constrains plants to a narrow area resulting in a
the incidence of invasive taxa 19 0.64 0.90 breakdown of the successional process (Ciccarelli et al.
(see ‘‘Materials and Methods’’ 20 0.65 0.88 2012).
Section for details) As mentioned above, various studies have shown dune
vegetation to be particularly sensitive to human trampling
(Andersen 1995; Lemauviel and Rozé 2003; Kerbiriou
Shifting dunes with A. arenaria (EC 2120) have been et al. 2008; Santoro et al. 2012; Farris et al. 2013; Fenu
traditionally considered indicative of dune systems char- et al. 2013b). The main adverse effect of trampling is the
acterized by heavy sand deposition (Ranwell 1972; Huiskes partial or total destruction of vegetation; while another
1979; Lubke and Hertling 2001). However, in the present effect is soil compaction, which increases soil density and
study this habitat was predominant in transects affected by decreases porosity, an unfavorable condition especially for
erosion, where the coastal processes have canceled the plant communities of embryonic and mobile dunes that are
previous plant communities of the driftline and the adapted to substrate instability and sand burial (Maun
embryonic dunes, leaving the mobile dunes with A. are- 2009). Fortunately it has been shown that dune plant
naria exposed to the seaside, as suggested by Ciccarelli communities are able to recover in a relatively short time if
et al. (2012). The disruption of mobile dunes may be a the disturbance factor (e.g., trampling) is removed, because
dramatic phenomenon because it could promote strong of the pioneer features that are typical of coastal dune
accumulation of sand in interior areas causing the occur- plants (Santoro et al. 2012).
rence of vegetation types that are normally typical of the In accordance with Fenu et al. (2013a), the results of this
foredunes and a decrease of plant communities typical of research confirm that distance from the coastline and total
the interdunal spaces and fixed dunes (Lomba et al. 2008). length of transects do not significantly affect plant com-
The C. maritimae habitat (EC 2210) was quite repre- munity composition along the Italian coasts, differently
sented in transects, showing often a mosaic pattern with from what was reported by Guara-Requena (1989) for
Malcolmietalia grasslands (EC 2230), which is rare and Spain. As suggested by Fenu et al. (2013a), plant com-
vulnerable in Tuscany (REpertorio NAturalistico TOscano munity distribution along the gradient coast-to-inland
(RENATO) 2005). cannot be sufficiently explained only by the distance from
The PCO ordination was mainly associated with the the shoreline at local scale.
coast-to-inland sequence of vegetation, and evidenced a Analysis of plant diversity of transects by means of
high correlation with four plant taxa—A. arenaria, E. diversity indices confirmed the low usefulness of N to
farctus, H. stoechas, and V. fasciculata—that are the main assess human impact on coastal dunes, as evidenced in
diagnostic species of Italian EU coastal habitats (Fig. 2). C. other studies on Mediterranean coastal sand dunes (De
maritimae habitat (EC 2210) and Malcolmietalia grass- Luca et al. 2011; Attorre et al. 2013). The small variability
lands (EC 2230) showed a certain degree of overlapping of N was due to the low incidence of alien taxa in the study
that could be explained by the less selective conditions of area, where I find only A. donax, C. cesatiana, E.

123
202 Environmental Management (2014) 54:194–204

angustifolia, E. canadensis, O. biennis, and X. orientale necessary at increasingly high costs. A better alternative is
subsp. italicum, which is considered as an invasive plant of to preserve these natural ecosystems and use the many
dunes although it belongs to the Italian flora. As seen in De ecosystem services in a more rational manner (Pérez-
Luca et al. (2011) and in Attorre et al. (2013), the appli- Maqueo et al. 2013).
cation of the NHDune diversity index may be an important
tool to quantify species richness on coastal dunes, although Acknowledgments The author thanks the Migliarino-San Rossore-
Massaciuccoli Regional Park and the Maremma Regional Park for
NHDune better discriminates between sites with the same permission to conduct this research, and the Italian Ministry of Uni-
taxa but with different habitat coverage. Unfortunately, in versity and Research for providing financial support. Thank are also
this study no significant correlations have been highlighted due to Francesca Marcacci and Paolo Montagnini for help during field
with single disturbance factors, probably due to the fact sampling. Lastly, the author thanks the Editor-in-Chief and two
anonymous reviewers for their constructive comments that improved
that in our study area plant communities belonging to the the previous versions of the paper.
same habitat are characterized by a certain degree of het-
erogeneity in species composition.
Given the present findings, several important consider-
ations regarding the management of coastal dune systems References
of the protected areas studied should be made. First, ero-
sion is the most important factor of disturbance that causes Acosta A, Ercole S, Stanisci A, Pillar VDP, Blasi C (2007) Coastal
structural alterations of the typical spatial pattern of dune vegetation zonation and dune morphology in some Mediterra-
nean Ecosystems. J Coast Res 23:1518–1524
plant communities. The most common approaches to pro- Acosta A, Carranza ML, Izzi CF (2009) Are there habitats that
tecting an eroding foredune are the following: restoring the contribute best to plant species diversity in coastal dunes?
beach/foredune to semi-stable by arresting sand movement Biodivers Conserv 18:1087–1098
(e.g., with erection of fencing together with planting dune- Akaike H (1974) A new look at the statistical model identification.
IEEE Trans Automat Control 19:716–723
forming species such as A. arenaria), promoting accretion Andersen UV (1995) Resistance of Danish coastal vegetation types to
by sand nourishment or curtailing those activities that human trampling. Biol Conserv 71:223–230
reduce the sediment supply to the beach (Doody 2013). Anfuso G, Pranzini E, Vitale G (2011) An integrated approach to
During recent decades, the building of protective structures coastal erosion problems in northern Tuscany (Italy): littoral
morphological evolution and cell distribution. Geomorphology
(e.g., groynes, seawalls, jetties) along the coastal traits 129:204–214
affected by erosion has been a solution only for urgent Arrigoni PV, Viegi L (2011) La flora vascolare esotica spontaneizzata
local problems, because erosion processes shifted the della Toscana. Regione Toscana, Firenze
downdrift of harbors and other engineering structures. In Attorre F, Maggini A, Di Traglia M, De Sanctis M, Vitale M (2013) A
methodological approach for assessing the effects of disturbance
fact, as evidenced by Anfuso et al. 2011, in northern factors on the conservation status of Mediterranean coastal dune
Tuscany (Italy) the human-made protective structures systems. Appl Veg Sci 16:333–342
prevented coastal retreat at local scale, but caused down- Biondi E, Blasi C, Burrascano S, Casavecchia S, Copiz R, Del Vico
drift erosion, modifying sediment alongshore flow and E, Galdenzi D, Gigante D, Lasen C, Spampinato G, Venanzoni
R, Zivkovic L (2009) Manuale Italiano di Interpretazione degli
increasing wave-related processes. These phenomena Habitat della Direttiva 92/43/CEE. Società Botanica Italiana—
caused the degradation of several coastal stretches which Ministero dell’Ambiente e della tutela del Territorio e del Mare,
require continuous management actions. Direzione Protezione della Natura. Available at http://vnr.unipg.
Second, trampling and presence of paths, connected with it/habitat/. Accessed 02 Sept 2013
Bjornstad ON (2013) Ncf: spatial nonparametric covariance func-
human recreational activities, appear to have a detrimental tions. R package version 1.1-5. http://onb.ent.psu.edu/onb1/R.
impact on sand dunes. To reduce human pressure managers Accessed 09 Dec 2013
should narrow or reroute pedestrian traffic through the Borcard D, Legendre P, Drapeau P (1992) Partialling out the spatial
installation of footbridges, and the use of appropriate fences component of ecological variation. Ecology 73:1045–1055
Brown AC, McLachlan A (2002) Sandy shore ecosystems and the
to allow aeolian sand transport and drift. These permit a threats facing them: some predictions for the year 2025. Environ
more natural dynamic dune form to develop (Doody 2013). Conserv 29:62–77
The economic benefits that society receive from the Cabeza M, Moilanen A (2001) Design of reserve networks and the
natural functioning of the beach and coastal dunes are very persistence of biodiversity. Trends Ecol Evol 16:242–248
Carboni M, Carranza ML, Acosta A (2009) Assessing conservation
high, especially from the point of view of protection, rec- status on coastal dunes: a multiscale approach. Landsc Urban
reation, and esthetic factors. The paradox is that, because Plan 91:17–25
of the strong interest in these ecosystems, coasts are often Ciccarelli D, Bacaro G, Chiarucci A (2012) Coastline dune vegetation
over-exploited in order to obtain short-term benefits from dynamics: evidence of no stability. Folia Geobot 47:263–275
Conti F, Abbate G, Alessandrini A, Blasi C (2005) An annotated
recreation and their natural beauty. Such over-exploitation checklist of the Italian vascular flora. Palombi Editori, Roma
results in a degraded and even lost natural ecosystem. In Conti F, Alessandrini A, Bacchetta G, Banfi E, Barberis G, Bartolucci
consequence, restoration efforts are becoming increasingly F, Bernardo L, Bonacquisti S, Bouvet D, Bovio M, Brusa G, Del

123
Environmental Management (2014) 54:194–204 203

Guacchio E, Foggi B, Frattini S, Galasso G, Gallo L, Gangale C, Gornish ES, Miller TE (2010) Effects of storm frequency on dune
Gottschlich G, Grünanger P, Gubellini L, Iriti G, Lucarini D, vegetation. Global Change Biol 16:2668–2675
Marchetti D, Moraldo B, Peruzzi L, Poldini L, Prosser F, Gower JC (1966) Some distance properties of latent root and vector
Raffaelli M, Santangelo A, Scassellati E, Scortegagna S, Selvi F, methods used in multivariate analysis. Biometrika 53:325–338
Soldano A, Tinti D, Ubaldi D, Uzunov D, Vidali M (2007) Grunewald R, Schubert H (2007) The definition of a new plant
Integrazioni alla checklist della flora vascolare italiana. Nat diversity index ‘‘H’dune’’ for assessing human damage on coastal
Vicentina 10:5–74 dunes—derived from the Shannon index of entropy H’. Ecol
Coombes EG, Jones AP, Sutherland WJ (2008) The biodiversity Indic 7:1–21
implications of changes in coastal tourism due to climate change. Gruppo Nazionale per la ricerca sull’ambiente costiero (2006) Lo
Environ Conserv 35:319–330 stato dei litorali italiani. Studi Costieri 10:1–174
Cori B (1999) Spatial dynamics of Mediterranean coastal regions. Guara-Requena M (1989) La influencia de la distancia al mar en la
J Coast Conserv 5:105–112 distribucion de la flora de las dunas del Cabo de Gata. Acta Bot
Council of European Communities (1992) Council Directive 92/43/ Malacit 14:151–159
EEC of 21 May 1992 on the conservation of natural habitats and Huiskes AH (1979) Biological Flora of the British Isles. Ammophila
of wild fauna and flora. Off J Eur Communities 35:7–50 arenaria (L.) Link. J Ecol 67:363–382
Council of European Communities (1997) Formulario informativo sui Kerbiriou C, Leviol I, Jiguet F, Julliard R (2008) The impact of
siti proposti per l’inserimento nella rete Natura 2000, Annex E. human frequentation on coastal vegetation in a biosphere
Gazzetta ufficiale delle Comunità europee L 107/1 reserve. J Environ Manag 88:715–728
D’Ortenzio F, Iudicone D, de Boyer Montegut C, Testor P, Antoine King CAM (1975) Introduction to physical and biological oceanog-
D, Marullo S, Santoreli R, Madec G (2005) Seasonal variability raphy. Edward Arnold, London
of the mixed layer depth in the Mediterranean Sea as derived Kutiel P (2001) Conservation and management of the Mediterranean
from in situ profiles. Geophys Res Lett 32:L12605 coastal sand dunes in Israel. J Coast Conserv 7:183–192
De Luca E, Novelli C, Barbato F, Menegoni P, Iannetta M, Nascetti G Kutiel P, Zhevelev H, Harrison R (1999) The effect of recreational
(2011) Coastal dune systems and disturbance factors: monitoring impacts on soil and vegetation of stabilised coastal dunes in the
and analysis in central Italy. Environ Monit Assess 183:437–450 Sharon Parks, Israel. Ocean Coast Manag 42:1040–1060
Doing H (1985) Coastal fore-dune zonation and succession in various Lavorel S, Canadell J, Rambal S, Terradas J (1998) Mediterranean
parts of the world. Vegetatio 61:65–75 terrestrial ecosystems: research priorities on global change
Doody JP (2013) Sand dune conservation, management and restora- effects. Glob Ecol Biogeogr Lett 7:157–166
tion. Springer, Heidelberg Legendre P, Anderson MJ (1999) Distance-based redundancy
Drius M, Malavasi M, Acosta ATR, Ricotta C, Carranza ML (2013) analysis: testing multispecies responses in multifactorial ecolog-
Boundary-based analysis for the assessment of coastal dune ical experiments. Ecol Monogr 69:1–24
landscape integrity over time. Appl Geogr 45:41–48 Legendre P, Fortin M-J (1989) Spatial pattern and ecological analysis.
EEA (1999) Coastal and marine zones: environment at the turn of the Vegetatio 80:107–138
century. Chap 3/14. EEA, Copenhagen Lemauviel S, Rozé F (2003) Response of three plant communities to
Falcucci A, Maiorano L, Boitani L (2007) Changes in land-use/land- trampling in a sand dune system in Brittany (France). Environ
cover patterns in Italy and their implications for biodiversity Manag 31:227–235
conservation. Landsc Ecol 22:617–631 Lomba A, Alves P, Honrado J (2008) Endemic sand dune vegetation
Farris E, Pisanu S, Ceccherelli G, Filigheddu R (2013) Human of the Northwest Iberian Peninsula: diversity, dynamics, and
trampling effects on Mediterranean coastal dune plants. Plant significance for bioindication and monitoring of coastal land-
Biosyst 147:1043–1051 scapes. J Coast Res 24:113–121
Fenu G, Cogoni D, Ferrara C, Pinna MS, Bacchetta G (2012) Lubke RA, Hertling UM (2001) The role of European marram grass
Relationships between coastal sand dune properties and plant in dune stabilization and succession near Cape Agulhas, South
community distribution: the case of Is Arenas (Sardinia). Plant Africa. J Coast Conserv 7:171–182
Biosyst 146:586–602 Malavasi M, Santoro R, Cutini M, Acosta ATR, Carranza ML (2013)
Fenu G, Carboni M, Acosta A, Bacchetta G (2013a) Environmental What has happened to coastal dunes in the last half century? A
factors influencing coastal vegetation pattern: new insights from multitemporal coastal landscape analysis in Central Italy. Landsc
the Mediterranean basin. Folia Geobot 48:493–508 Urban Plan 119:54–63
Fenu G, Cogoni D, Ulian T, Bacchetta G (2013b) The impact of Margules CR, Pressey RL (2000) Systematic conservation planning.
human trampling on a threatened coastal Mediterranean plant: Nature 405:243–252
the case of Anchusa littorea Moris (Boraginaceae). Flora Martı́nez ML, Maun AM, Psuty N (2004) The fragility and
208:104–110 conservation of the world’s coastal dunes: geomorphological,
Feola S, Carranza ML, Schaminéè JHJ, Janssen JAM, Acosta ATR ecological and socioeconomic perspectives. In: Martı́nez ML,
(2011) EU habitat of interest: an insight into Atlantic and Psuty N (eds) Coastal dunes: ecology and conservation.
Mediterranean beach and foredunes. Biodivers Conserv Springer, Heidelberg, pp 355–369
20:1457–1468 Martı́nez ML, Gallego-Fernández JB, Garcı́a-Franco JG, Moctezuma
Forey E, Chapelet B, Vitasse Y, Tilquin M, Touzard B, Michalet R C, Jiménez CD (2006) Assessment of coastal dune vulnerability
(2008) The relative importance of disturbance and environmen- to natural and anthropogenic disturbances along the Gulf of
tal stress at local and regional scales in French coastal sand Mexico. Environ Conserv 33:109–117
dunes. J Veg Sci 19:493–502 Maun MA (2009) The biology of coastal sand dunes. Oxford
Frederiksen L, Kollmann J, Vestergaard P, Bruun HH (2006) A University Press, Oxford
multivariate approach to plant community distribution in the Mendoza-González G, Martı́nez ML, Lithgow D, Pérez-Maqueo O,
coastal dune zonation of NW Denmark. Phytocoenologia Simonin P (2012) Land use change and its effects on the value of
36:321–342 ecosystem services along the coast of the Gulf of Mexico. Ecol
Garcı́a-Mora MR, Gallego-Fernández JB, Williams AT, Garcı́a-Novo Econ 82:23–32
F (2001) A coastal dune vulnerability classification. A case study Miller TE, Gornish ES, Buckley H (2010) Weather and coastal
of the SW Iberian Peninsula. J Coast Res 17:802–811 vegetation: effects of storms and drought. Plant Ecol 206:97–104

123
204 Environmental Management (2014) 54:194–204

Muñoz Vallés S, Gallego Fernández JB, Dellafiore CM (2011) Dune Rivas-Martinez S (2005) Notions on dynamic-catenal phytosociology
vulnerability in relation to tourism pressure in central Gulf of as a basis on landscape science. Plant Biosyst 139:135–144
Cádiz (SW Spain), a case study. J Coast Res 27:243–251 Santoro R, Jucker T, Prisco I, Carboni M, Battisti C, Acosta ATR
Økland RH (2003) Partitioning the variation in a plot-by-species data (2012) Effects of trampling limitation on coastal dune plant
matrix that is related to n sets of explanatory variables. J Veg Sci communities. Environ Manag 49:534–542
14:693–700 Shannon CE, Weaver W (1949) A mathematical theory of commu-
Oksanen J, Blanchet FG, Kindt R, Legendre P, Minchin PR, O’Hara nication. University of Illinois, Champaign
RB, Simpson GL, Solymos P, Stevens MHH, Wagner H (2011) Stallins JA (2002) Dune plant species diversity and function in two
Vegan: community ecology package. R package version 2.0-2. barrier island biogeomorphic systems. Plant Ecol 165:183–196
http://cran.r-project.org/package=vegan. Accessed 02 Sept 2013 R Development Core Team (2011) R: A language and environment
Peres-Neto PR, Legendre P, Dray S, Borcard D (2006) Variation for statistical computing. R Foundation for Statistical Comput-
partitioning of species data matrices: estimation and comparison ing. Vienna, Austria. http://cran.r-project.org/. Accessed 02 Sept
of fractions. Ecology 87:2614–2625 2013
Pérez-Maqueo O, Martı́nez ML, Lithgow D, Mendoza-González G, ter Braak CJF (1986) Canonical correspondence analysis: a new
Feagin RA, Gallego-Fernández JB (2013) The coasts and their eigenvector technique for multivariate direct gradient analysis.
costs. In: Martı́nez ML, Gallego-Fernández JB, Hesp PA (eds) Ecology 67:1167–1179
Restoration of coastal dunes. Springer, Heidelberg, pp 289–304 Van der Maarel E (2003) Some remarks on the functions of European
Pinna M (1985) Contributi di Climatologia. Mem Soc Geogr Ital coastal ecosystems. Phytocoenologia 33:187–202
39:147–167 Van der Maarel E, Van der Maarel-Versluys M (1996) Distribution
Provoost S, Ampe C, Bonte D, Cosyns E, Hoffmann M (2004) and conservation status of littoral vascular plant species along
Ecology, management and monitoring of grey dunes in Flanders. the European coasts. J Coast Conserv 2:73–92
J Coast Conserv 10:33–42 Weyl PK (1970) Oceanography. An introduction to the marine
Psuty N (2004) The coastal foredune: a morphological basis for environment. Wiley, New York
regional coastal dune development. In: Martı́nez ML, Psuty N Wiedemann AM, Pickart A (2004) Temperate zone coastal dunes. In:
(eds) Coastal dunes: ecology and conservation. Springer, Hei- Martı́nez ML, Psuty N (eds) Coastal dunes: ecology and
delberg, pp 11–27 conservation. Springer, Heidelberg, pp 53–65
Ranwell DS (1972) Ecology of salt marshes and sand dunes. Podani J (2007) Analisi ed esplorazione multivariata dei dati in
Chapman and Hall, London ecologia e biologia. Liguori Editore, Napoli
Rapetti F (2003) Il clima. In: AAVV (ed) Atlante tematico della Yoccoz NG, James DN, Boulinier T (2001) Monitoring of biological
Provincia di Pisa. Pacini Editore, Pisa, pp 16–17 diversity in space and time. Trends Ecol Evol 16:446–453
REpertorio NAturalistico TOscano (RENATO) (2005) Identificazione
dei target di biodiversità in Toscana. Regione Toscana, Firenze

123