Journal of Coastal Research      22      5     1145–1157      West Palm Beach, Florida     September 2006




Assessment of Damages from Recreational Activities on
Coastal Dunes of the Southern Baltic Sea
Ralf Grunewald

Institute of Ecology
University of Rostock
Albert-Einstein-Strasse 3
18051 Rostock, Germany
ralf.grunewald@biologie.uni-rostock.de

             ABSTRACT
             GRUNEWALD, R., 2006. Assessment of damages from recreational activities on coastal dunes of the southern Baltic
             Sea. Journal of Coastal Research, 22(5), 1145–1157. West Palm Beach (Florida), ISSN 0749-0208.

             A study of the changes in natural phytodiversity that can be linked to recreational activities on tertiary dunes (grey
             dunes) is presented. Ten neighboring areas on the Islands of Wolin (Poland) and Usedom (Germany), representing all
             major disturbance factors (holiday resorts, campgrounds, parking), were chosen with two appropriate reference areas
             from Usedom and Wolin Island (Poland) that are situated in remote parts of the coast and showed no permanent
             signs of human disturbance. The results of recreational activities, especially trampling, eutrophication, and the neigh-
             boring effects of nearby gardens, parks, or fallow land, were assumed to be the main factors influencing natural dunal
             phytodiversity. Mechanical stress from trampling could result in a decrease in phytodiversity, whereas rising nutrient
             levels (e.g., fecal deposits), as well as neighboring effects from urban development and a growing species pool, could
             lead to an increase in diversity, including the occurrence of generalist or ruderal species not typical of unmodified
             dunes. Hence, changes within natural dunal phytodiversity depend on character and intensity of recreational activi-
             ties. Results show that the location of the beach largely influences visitor numbers and hence the level of disturbance.
             At holiday resorts, damage to dunes can be low, even though tourist numbers on the beach are very high, if the dunes
             can be overlooked from the main promenade. Visitors are reluctant to trespass onto the dunes if they are being
             watched by many people. Nevertheless, areas without major signs of human disturbance were mostly found on remote
             dunes that were only accessible by foot or bicycle. Application of established phytodiversity indices (e.g., H ) showed
             drawbacks in detecting the different types of disturbance; however, sensitivity is improved by a modified index
             (Hdune).

             ADDITIONAL INDEX WORDS: Trampling, plant diversity, diversity index, eutrophication, Pomeranian bight, Baltic
             Sea, Wolin, Usedom.




           INTRODUCTION                     seems more useful because more subtle changes in nutrient
                                      availability or neighboring effects from disturbances nearby
  Several studies have evaluated the effects of anthropogenic
                                      might lead to significant changes in species composition and
activities on coastal ecosystem diversity (e.g., BOGAERT and
                                      community structure, as was also described by GRIME (2002)
LEMEUR, 1995; BUNDESAMT FUR NATURSCHUTZ, 1997; HYL-
                  ¨
                                      or in the ‘‘intermediate disturbance hypothesis’’ (CONNELL,
GAARD and LIDDLE, 1981; ISERMANN, 1995; JESCHKE, 1985;
                                      1978). PICKETT and WHITE (1985) define disturbance as ‘‘any
LEMAUVIEL and ROZE, 2003; LIDDLE, 1975; LIDDLE and
              ´
                                      relatively discrete event in time that disrupts ecosystem,
GREIGH-SMITH, 1975; PETERS and POTT, 1999; RODGERS and
                                      community or population structure and changes resources,
PARKER, 2003).
                                      substrate availability, or the physical environment.’’
  The aim of this study is to evaluate disturbances linked to
                                       Generally, coastal ecosystems and their associated biodi-
diffuse activities, i.e., recreational activities on coastal dunes.
                                      versity are described as especially sensitive to human distur-
The hypothesis was that natural plant diversity could be used
                                      bance such as trampling and eutrophication (JESCHKE, 1985;
as an indicator of human disturbance. Disturbance is defined
                                      KNAPP, 1996; KUTIEL, ZHEVELEV, and HARRISON, 1999; LID-
in this study as any change in plant cover or species compo-
                                      DLE, 1975; LIDDLE and GREIGH-SMITH, 1975; MCDONNELL,
sition that can be linked to human (recreational) activities.
                                      1981; VON NORDHEIM and BOEDEKER, 1998). However, ir-
This definition is much broader than that used by G RIME
                                      respective of more general descriptions, only a few studies
(2002), who restricts disturbance to ‘‘mechanisms which limit
                                      deal with quantification of diffuse disturbances like tram-
the plant biomass by causing its partial or total destruction.’’
                                      pling or eutrophication and the methods used differ largely.
Instead, a different definition by P ICKETT and WHITE (1985)
                                       Nevertheless, some general conclusions on the effects of hu-
                                      man disturbance on the phytodiversity of dunes can be
DOI:10.2112/05-0464.1 received 15 February 2005; accepted in revi-
                                      drawn. Commonly, a decrease of diversity (both species rich-
sion 11 November 2005.
 The work was financed by the German Federal Environmental         ness and evenness; see, e.g., BAUR and EHRHARDT, 1995; VAN
Foundation (DBU).                             DER MAAREL, 1971) is believed to be the result of any sub-
1146                              Grunewald


stantial anthropogenic action. A study on coastal dunes in
Finland (HELLEMAA, 1998) showed a degradation of flora by
comparing white dunes (secondary dunes) influenced by hu-
man activities (mostly trampling). MCDONNELL (1981) also
showed that all studied levels of trampling intensity will re-
duce species diversity, mostly plant cover. On the other hand,
studies by PIOTROWSKA (1979, 1989) or PIOTROWSKA and
STASIAK (1982) revealed an increase in species richness, es-
pecially in the older dune formations (grey and brown dunes,
i.e., tertiary dunes), which was inter alia explained by effects
from eutrophication and a more intensive land use adjacent
to the dunes (neighboring effects). Investigations by RODG-
ERS (2002) on Atlantic barrier Islands in the US state of
Georgia also showed the effect human disturbance can have
on species richness and increasing occurrences of generalists
and alien species. ROGERS (2002) and RODGERS and PARKER
(2003) did not find a generally increased species richness of
alien species with respect to human disturbance (they did
find higher coverage values for alien species in disturbed
sites) but also did not investigate dunes in the vicinity of       Figure 1. (A) Typical transect through a heavily changed beach and
                                     dune system along the German coast of the southern Baltic Sea. (B) Tran-
holiday resorts. In those areas, neighboring effects from hol-
                                     sect through an almost unchanged beach and dune system (except for
iday resorts or harbor installations might also influence spe-      forest plantation) with a succession of dunal ridges and depressions of
cies diversity because the species pool of these sites is larger     different height and age (figure adapted and taken from Grunewald and
than at isolated sites (species pool hypothesis, PARTEL et al.,
                          ¨           Schubert (2005).
1996). Finally, the intermediate disturbance hypothesis
might explain a rise in diversity (CONNELL, 1978) as well. So
character and intensity of disturbance (trampling or eutro-
                                     cession, together with pedogenic processes (e.g., leaching of
phication) are important for the effect caused. Because all
                                     calcium or the accumulation of organic matter from white
studies have shown that natural phytodiversity is affected by
                                     dunes over grey dunes to brown dunes), will slowly dampen
human activity, the use of biodiversity indices seemed a rea-
                                     the conditions by building up a humus layer, accumulating
sonable approach for quantification, and their use should be
                                     nutrients, and storing water (ELLENBERG, 1996; HESP,
tested in this study. Thus, in this study, I try to define a
                                     1991). However, with respect to daily temperature amplitude,
measure on the basis of phytodiversity for quantification of
                                     water availability, and nutrient levels, tertiary dunes are still
the different types of human activity on tertiary dunes.
                                     extreme habitats. Plant species show a number of different
  Along the southern Baltic Sea coast, urban development in
                                     adaptations to cope with these conditions and can be regard-
combination with recreational activities and coastal defense
                                     ed as specialized dune species forming typical dune commu-
measures has changed and altered the shape and plant spe-
                                     nities (HESP, 1991). Among others are many drought-resis-
cies composition of most low-lying beaches with dunes (IS-
                                     tant species of mosses and lichens, many of which are espe-
ERMANN, 1997; JESCHKE, 1985). However, along prograding
                                     cially sensitive to trampling (ISERMANN, 1997).
coastlines, new dunes are constantly being formed, and the
dynamic beach and dune ecosystem could be seen as one of
                                             MATERIAL AND METHODS
the few primary habitats still remaining in Central Europe
(LANDESAMT FUR FORSTEN UND GROSSSCHUTZGEBIETE
         ¨                             The data used in this study were gathered in 2002 and
MECKLENBURG-VORPOMMERN and NATIONALPARKAMT                2003 on tertiary dunes along the coasts of Usedom (Germany)
VORPOMMERSCHE BODDENLANDSCHAFT, 2002). Figure 1             and Wolin Island (Poland) (Figure 2). Both islands were se-
shows one transect through an almost undisturbed dune sys-        lected because they share the same abiotic and biotic condi-
tem and one through a typically altered dune system along        tions along the southern Baltic Sea, so that any changes in
the German Baltic Coast. The definition of the different dune       plant composition or community structure are most likely not
types is based on the classification used in the Red List of       linked to differences in habitat conditions. The coast is char-
biotopes and biotope complexes of the Baltic Sea, Belt Sea,       acterized by alternating Pleistocene moraine cores (under
and Kattegat (VON NORDHEIM and BOEDEKER, 1998), which          abrasion) and Holocene marine sediments (sand accumula-
differs from other definitions (e.g., HESP, 1991).            tion), which form sandy spits or barrier islands or enclose
  Because constantly blowing winds, moving sand (sand          lagoons. Material from the moraine sediments is abraded,
blasting), salt, and heavy storms inflict either continuous or      sorted, transported by the sea, and finally accumulated along
abrupt changes, only specialized plants are able to survive       low-lying, prograding coastlines where sand, wind, and veg-
within the beach and primary or secondary dune habitats         etation interact to form beaches and dunes. The dunes stud-
(HESP, 1991). This study focuses on tertiary dunes, which        ied are approximately only a few up to a hundred years old
provide some shelter from sand blasting and in which salt        (JANKE, 1971; KLIEWE and RAST, 1979). The temperate cli-
water influence is minimal (H ESP, 1991). The vegetation suc-       matic conditions are very similar (yearly mean 7.5 C, year-



                       Journal of Coastal Research, Vol. 22, No. 5, 2006
                             Tourism and Plant Diversity on Dunes                             1147


                                        sites should therefore share the same potential for dune veg-
                                        etation.
                                          Usedom and Wolin Islands are located in the Polish-Ger-
                                        man transborder region (Figure 2), and tourism is one of the
                                        few thriving industries (SEIDEL, 2001). Growing numbers of
                                        visitors are expected in Poland, whereas numbers are already
                                        high in Germany. Because large stretches along the Wolin
                                        coast are only accessible by foot or bicycle, these areas so far
                                        remain in a fairly undisturbed ecological state, and coastal
                                        recreation in northwestern Poland is concentrated mostly
                                        around holiday resorts.
                                          Altogether 12 different sites along the coast were selected
                                        for this study (Figure 2 and Table 1). The level of accessibility
                                        and control was used to choose sites because this can be seen
                                        as one indication of the intensity of recreational use. In ad-
                                        dition, tourist counts were carried out along the northern
Figure 2. The Pomeranian Bight with study sites around Karlshagen,
Trassenheide, and Heringsdorf on Usedom Island (see description in Ta-     part of Usedom Island to verify the assumption that remote
ble 2) and on the Swina Gate Barrier (Wolin Island, Poland).          areas are being visited by fewer people than easily accessible
                                        sites, as was also shown by KAMMLER (2003) for other sites
                                        along the southern Baltic Sea coast. Site 1 (Figure 3) is sit-
ly precipitation 575 mm; MULLER, 2004; REINHARD, 1951,
                ¨                        uated on the central part of Wolin (Swina Gate Barrier) be-
                                              ˛        ´
                                        tween Miedzyzdroje and Swinoujscie, Poland, and is visited
                                                          ´
1962). The two islands are characterized by sub-Atlantic cli-
mate because both are strongly influenced by their position           by very few people (for a detailed description, see ŁABUZ and
                                                                  ´
                                        GRUNEWALD, 2007). Sites 2 and 3 are closer to Swinoujscie,  ´
between the two waterbodies of the Baltic Sea and the Szcze-
cin and Odra Lagoons. Coastal exposition changes only little          and tourists regularly use the area because parking lots are
between the sites (i.e., tertiary dunes), and westerly winds          close to the beach. In addition, site 3 is also close to the har-
prevail during most of the year. Storm surges with peak wind               ´
                                        bor of Swinoujscie (possible neighboring effects). Site 4 on
                                                 ´
speeds mostly come from northeasterly directions from No-            Usedom Island is situated in Heringsdorf; many tourists use
vember to February (HURTIG, 1957). The type of sand of             the area, but for daytime visitors, the beach is not easily ac-
which the dunes consist is also very similar in mineral com-          cessible. Furthermore, public toilets are available and rescue
position, as well as grain size (ISERMANN, 2001). All study           swimmers regularly control beach and dunes. Sites 5, 6, and


Table 1. Study sites and associated recreational activities.

            Study Site

  No.             Name                                Recreational Activity

   1       Wolin, central Barrier            Central part of Wolin Island (Swina Gate Barrier): very few visitors, isolated, no park-
                                 ing, no public transportation (see also: Łabuz and Grunewald, 2004)
   2             ´
           Wolin, near Swinoujscie
                    ´             Western part of Wolin Island (Swina Gate Barrier): high numbers of visitors, but a
                                 nearby bar overlooking the dunes ‘‘controls’’ number of visitors on the dunes
   3       Wolin, Harbor                Westernmost part of Wolin Island (Swina Gate Barrier): high numbers of visitors, but
                                        ´
                                 the tower of Swinoujscie harbor ‘‘controls’’ number of visitors on the dunes
                                            ´
   4       Heringsdorf                 Eastern part of Usedom Island. Very popular holiday resort, but no parking facilities
                                 for day tourists; public toilets available, lifeguards control beach and dunes
   5       Trassenheide I                Northern part of Usedom Island: small holiday resort; open dunes can be easily observed
                                 from beach and official dune crossings
   6       Trassenheide II               Northern part of Usedom Island: small holiday resort; open dunes can be easily observed
                                 from beach and official dune crossings
   7       Trassenheide, center             Northern part of Usedom Island: small holiday resort; open dunes can be easily observed
                                 from beach and official dune crossings as well as a beachside promenade
   8       Between Trassenheide and Karlsh-       Northern part of Usedom Island: very few visitors, isolated, no parking, no public trans-
           agen                     portation
   9       Campground                  Northern part of Usedom Island: adjacent to a large very popular campground south of
                                 Karlshagen, but no public parking available, so mostly campers use the beach
   10       Karlshagen, center              Northern part of Usedom Island next to the main square in the holiday resort of Karlsh-
                                 agen: very high numbers of tourists; dunes can be overlooked from the beach and official
                                 dune crossings
   11       Karlshagen-North               Northern part of Usedom Island just north of the site 10: trees, bushes, and shrubs
                                 block the view
   12       Parking                   Northern part of Usedom Island adjacent to a large parking lot north of Karlshagen:
                                 very popular; dune system is not as wide and complex as the other sites (fewer plots)




                           Journal of Coastal Research, Vol. 22, No. 5, 2006
1148                                 Grunewald




Figure 3. Study site 11: dunes within the holiday resort of Karlshagen showing clear signs of heavy disturbance from trampling. For color version of
this figure, see page 1173.




7 are near or within the small holiday resort of Trassenheide,        for the study. For each plot, all species were listed, and the
and public toilets are available. The number of tourists is          individual cover for each species was estimated according to
high, but the open dunes can be easily observed and con-           the adapted scale from Braun-Blanquet (adapted by BARK-
trolled from the beach or the official dune crossings (life-          MAN, DOING, and SEGAL, 1964). According to BRAUN-BLAN-
guards). Site 8 is a remote beach and dune area between the          QUET (1964) only homogeneous plots were studied, so that
holiday resorts of Karlshagen and Trassenheide. Like site 1,         any effects on diversity from changing habitat conditions
it is only accessible by foot or bicycle ( 2 km from the nearest       within one plot should be minimal (increasing habitat diver-
parking area or hotel). Site 9 is adjacent to a large camp-          sity  increasing species diversity). Even though epiphytic
ground south of Karlshagen; mostly campers use the beach,           lichens and mosses were not included, pedogenic lichens and
and public toilets are available. Sites 10 and 11 are used by         mosses were included in the study. Because both are on the
many holidaymakers and day tourists who use the nearby            same trophic level as vascular plants and make up a fair
parking areas. Public toilets are available, and site 10 is also       amount of plant diversity, they were included in the diversity
easily observable by rescue swimmers or people on the prom-          studies. PIOTROWSKA (1979) found that more than 50% of
enade. This is not the case for site 11, which is only about         phytodiversity on coastal dunes are cryptogams, which are
100 m north of site 10. At site 11, bushes and trees block the        known to be especially sensible to mechanical stress (ISER-
view from the beachside promenade and from the beach (Fig-          MANN, 1997) and should therefore be valuable indicators for
ure 4). Site 12 is north of the resort and adjacent to a large        changes in diversity. During summer drought situations, li-
parking lot. Many day tourists and overnight campers (also          chens and many mosses almost completely dry out and be-
illegal campers in the forested dunes) use the beach.             come fragile or brittle. Thus, any trampling on warm sunny
  The method of BARKMAN, DOING, and SEGAL (1964) and             days will result in heavy damage to the plants (ISERMANN,
BRAUN-BLANQUET (1964) was chosen to document plant spe-            1997).
cies presence and coverage (%) at 249 individual plots from           Heavy mechanical damages to the vegetation through
12 sites on Usedom and Wolin Islands. The minimum area            trampling, like large illegal footpaths or areas of bare sand
was determined, and a constant plot size of 16 m2 was chosen         with footprints, are clearly visible on tertiary dunes. There-



                         Journal of Coastal Research, Vol. 22, No. 5, 2006
                            Tourism and Plant Diversity on Dunes                         1149




Figure 4. Study site 1: dunes on Wolin Island showing no major signs of human disturbance. For color version of this figure, see page 1174.




fore, the following protocol to choose the individual plots at        a second approach focuses on quantitative changes that clear-
each study site was used. A minimum of three transects (from         ly can be linked to mechanical stress from trampling. Al-
young grey dunes to the older nonforested dunes; Figure 1)          though different diversity indices were initially calculated, an
were investigated at each of the 12 study sites: one covered         index (Hdune; GRUNEWALD, 2004; GRUNEWALD and SCHU-
the areas exhibiting the highest degree of damage within this         BERT, 2006) adapted from the Shannon and Wiener Index of
site (e.g., large illegal footpaths with bare sand, if present),       entropy (H , as described by HAEUPLER, 1982; MAGURRAN,
one had a representative level of damage within this site, and        1988) was used to indicate the main differences in diversity
one had the lowest level of damage for this study site. If some        between the dunes (Figure 5). The use of H revealed some
trees or shrubs were present, these sites were included, but         major drawbacks for this study because it strongly depends
the study mainly focused on the open dunes. For each study          on evenness and thus only measures the difference between
site, all plots from the three transects were pooled and the         the potential maximum and actual diversity regardless of the
groups were then used for analyzing and comparing the dif-          area covered by the species (Figure 6). HOBOHM and PETER-
ferent sites with each other. The number of plots per study          SEN (1999) and HAEUPLER (1982) already stated the strong
site can vary because the width of the dunes varies between          weight H puts on evenness. This makes a direct comparison
the different sites.                             of different plots at the same time or even monitoring the
  Because two main factors affecting diversity can be iden-         same plot over time difficult. Mechanical damages will lead
tified (eutrophication and trampling), but both have adverse          to a decrease in coverage (destruction on biomass), which
effects on the natural plant community (either increase or          could result in an increase in evenness (e.g., Figure 6: from
decrease of species diversity and plant cover), both factors         C2 to C1 or further to C3), which will lead to higher H values
must be dealt with separately. Growing levels of diffuse hu-         even though many species might be less abundant or even
man influence often correspond with increasing numbers of           become extinct.
generalist, alien, or both types of species (e.g., CORNELL,           The difference between H and Hdune is the definition of the
1999, species pool hypothesis; HILL, ROY, and THOMPSON,            sample size (Figures 5 and 6; GRUNEWALD and SCHUBERT,
2002; PARTEL et al., 1996). Therefore, one approach analyzes
     ¨                                  2005). As shown in GRUNEWALD and SCHUBERT (2005), the
qualitative changes in species diversity and plant cover and         sample size for H is not determined by the plot size only.



                          Journal of Coastal Research, Vol. 22, No. 5, 2006
1150                                 Grunewald


                                       Hdune. Furthermore, the strong weight H puts on the factor
                                       evenness is balanced by the factor coverage in Hdune. Even-
                                       ness is now mainly used to differentiate between communi-
                                       ties sharing roughly the same number of species and similar
                                       total coverage.
                                         Figure 5 indicates the different formula for Hdune-max (com-
                                       pared with the calculation on Hmax) that is needed for the
                                       calculation of evenness. Evenness with Hdune remains inde-
                                       pendent from both cover and species richness and can be used
                                       to analyze community structure in the same way evenness
                                       with H is used. This is also shown in Figure 6 and GRUNE-
                                       WALD and SCHUBERT (2005).
                                         As already described by PIOTROWSKA (1979) or RODGERS
Figure 5. The definition of Hdune and differences to the index H (Shan-    (2002), the number of species can also rise on tertiary dunes
non and Weaver, 1949). Figure adapted and taken from Grunewald and      as a result of eutrophication caused by tourism or general
Schubert (2005).                               development. Because other studies also showed that increas-
                                       ing levels of human influence will result in increasing num-
                                       bers of alien species (e.g., CORNELL, 1999, species pool hy-
Because the total number of individuals sampled is always           pothesis; GRIME, 2002; HILL, ROY, and THOMPSON, 2002;
set to 100%, the total number of individuals must be regarded         PARTEL et al., 1996), qualitative changes within the species
                                        ¨

as a second dimension of sample size, which is not kept con-         composition must be studied as well. GRIME (2002) states
stant with the use of H . It is an important difference wheth-        that ‘‘phosphorus and nitrogen are often the limiting resourc-
er plants cover 80% or only 30% of the constant sample area;         es in natural vegetation and several recent studies have
nevertheless, H will lose the information through normali-          shown that soil nutrient levels may play an important role
zation, and the index will then be dominated by evenness.           in determining a community’s invasibility.’’ In some areas on
This second normalization is omitted with the use of the           the studied dunes, many ruderal species were observed that
adapted index Hdune. In other words, the individual percent          were assumed to be a result of eutrophication; hence, the pre-
coverage, Pi-dune, is used to calculate the adapted Hdune, and        sented study tried to use the level of natural diversity (typical
sample size is only determined by unified plot size, which was         native dune species) as an additional indicator aiming at
kept constant at 16 m2 throughout the study. Because the           qualitative changes. To measure the level of natural diver-
second normalization of abundance relative to the total plant         sity, Hdune was calculated twice: first with all species included
cover is not made with Hdune, effects from a reduced plant          and then a second time excluding alien and ruderal species.
cover as a possible result of trampling are incorporated into         This second calculation could be interpreted as a measure-




Figure 6. Community models and diversity parameters for testing and comparing H and Hdune.




                         Journal of Coastal Research, Vol. 22, No. 5, 2006
                          Tourism and Plant Diversity on Dunes                        1151


ment of natural diversity of the different sites (Hdune-nat ). The     generation. Having two different main effects types—me-
ratio between the two values (Hdune-nat /Hdune ) could be seen as     chanical damage and eutrophication—classification should
the degree or level of natural diversity in the communities.        reflect both adequately. Therefore, classification was done
This indicator will be independent of species richness and         following the rationale that any disturbance so low in fre-
coverage, and a maximum value of one will only be reached         quency, amplitude, or both that does not change the system
if no alien or atypical species are found. Therefore, the species     within the time of observation must be ignored, and the sys-
found on all study sites were classified into typical native        tem still is considered to be ‘‘undisturbed’’ irrespective of no
dune species or species atypical for dunes or alien. Most of        longer being pristine. This rationale can be applied for both
the atypical species are also alien species to Poland or Ger-       mechanical action as well as eutrophication. If frequency or
many, but also, some native species alien to the dunal habitat       amplitude of the disturbance increases to an extent leading
were excluded from the calculation of Hdune-nat . The following      to a significant reaction of the system, this reaction might be
species were seen as being either atypical or alien to the         interpreted as a setback of the dunal succession from white
southern Baltic Sea dunes: Acer platanoides, Acer pseudopla-        dunes over grey to brown dunes (e.g., ELLENBERG, 1996;
tanus, Aira praecox, Artemisia vulgaris, Campylopos intro-         GRIME, 2002). This setback can be complete (i.e., back to bare
flexus, Conyza canadensis, Crataegus sp., Elaeagnus angusti-        sand) or incomplete to a pioneer stage or an intermediate
folia, Elaeagnus commutata, Prunus serotina, Rosa rugosa,         stage of succession. However, in both cases, the potential for
Sambucus nigra, Sedum sempervivum, Solidago virgaurea,           regeneration would be quite high. Restoration would take
Symphoricarpos albus, and Tortula ruraliformis. The classi-        place after switching off the effect and would follow a typical
fication was based on habitat requirements taken from the          dunal primary succession line, as might be expected for most
literature (ELLENBERG et al., 1992; HAEUPLER and MUER,           mechanical actions. So for all but the most severe mechanical
2000). Classification of the moss T. ruraliformis is based on        damage (heavy eutrophication or remobilization of the dune
personal experience, as well as on a study by BERG,            because of large areas of bare sand), such disturbances could
SCHRAMM, and DIEMINGER (unpublished study).                be regarded as an intermediate effect: the system would be
  In addition to the analysis of species diversity, so-called       regarded as good (minor disturbances) or disturbed (moder-
‘‘ecological indicator values’’ (ELLENBERG et al., 1992) were       ately disturbed) depending on the disturbance intensity. The
calculated as well. On the basis of decades of field studies        next higher level of disturbance would be an irreversible dis-
and vegetation analysis, Ellenberg developed a method to use        tortion leading to an environment changed so that succession
typical habitat conditions of plant species to assign ecological      would follow another line. Such changes would be regarded
indicator values for different parameters to the species. These      as ‘‘heavily disturbed’’ (e.g., heavy eutrophication, leading to
were developed for the parameters light (L), moisture (F),         a ruderalization of the vegetation; heavy trampling could lead
continentality/climate (K), temperature (T), nitrogen (N),         to a remobilization of the sand and parts of the dune). The
chemical reaction or soil acidity (R), and salinity (S). In this      worst case would be complete destruction and annihilated
study, the parameter N for nutrient conditions was calculated       restoration ability for either line of succession for a substan-
and used for the analysis of possible effects from eutrophi-        tial time period. Such changes would then be rated ‘‘very
cation. The maximum value of 9 will only be reached ‘‘in ex-        heavy damages and irreversibly substantially disturbed.’’ Ta-
tremely rich situation, such as cattle resting places or near       ble 2 gives an overview of the classification system.
polluted rivers,’’ and a value of 1 will only be reached in ‘‘sites      The method was tested along a gradient of different forms
extremely poor in available nitrogen’’ (ELLENBERG et al.,         and levels of recreational activities. Nine different study sites
1992, p. 251). Even though the ecological indicator values         on Usedom Island and three sites on Wolin Island were used
cannot be used as substitutes for exact measurements, they         to conduct a dune survey of tertiary dunes with presumably
do have the advantage of providing information on habitat         different levels of disturbance (Figure 2 and Table 1). Along
conditions for longer time periods because mostly perennials        Usedom Island, only one site was found that was remote and
are used for their calculation. For the nitrogen parameter,        thus not visited by many people. Because no signs of per-
the indicator will only reflect the presence of available nitro-      manent human influence were found, this site was used to-
gen, which is one advantage compared with some chemical          gether with the site on the central Wolin coast, which also
methods. The information is also integrated over the com-         did not show any clear signs of permanent human influence,
plete study area because average values are calculated on the       as reference areas or the reference status representing the
basis of all species present in a plot, and results have shown       class ‘‘undisturbed.’’ Because it is not possible to completely
significant correlations to measured data (E LLENBERG et al.,        prove that these areas have not changed from human activ-
1992). Abundance of species was not used to weight the cal-        ities and because minimal changes will remain unnoticed, the
culation because ELLENBERG et al. (1992) cited several stud-        classification system must be seen as subjective to a certain
ies that have shown that the more simple qualitative method        degree.
often yields better estimates of habitat conditions. This is ex-       In a first test, all plots were analyzed by nonparametric
plained inter alia by the low abundances that important in-        multidimensional scaling (MDS). Therefore, Bray-Curtis dis-
dicator plants with a high level of specialization on specific       similarity was applied on all plots with species and species
habitat conditions show in their preferred habitats.            abundance as variables. In a second step, the different plots
  For classification of the level of human-induced changes, a       were grouped into the 12 study sites and then analyzed and
system of five classes was used, defined according to both          compared with a Kruskal-Wallis one-way analysis performed
deviation from the natural state as well as potential for re-       on ranks and a following pairwise multiple comparison test



                        Journal of Coastal Research, Vol. 22, No. 5, 2006
1152                                  Grunewald


Table 2. A classification and evaluation system for coastal tertiary dunes.

      Ecological State                              Verbal Description

Undisturbed                   No signs of permanent human-induced changes
Only minor disturbances             Few typical species are missing, some signs of eutrophication, moderate signs of trampling, which could
                         be restored within a few years if disturbance is stopped
Medium disturbances               A number of alien and generalist species, obvious signs of trampling and eutrophication (some atypical
                         dune species present), but no large areas without vegetation
Heavy disturbances                Many alien and generalist species, heavy eutrophication and/or heavy trampling, with large areas with-
                         out vegetation; neighboring effects from nearby disturbances
Very heavy damages and irreversibly       Housing, construction, remobilization of the dune, covering dune sands with nutrient-rich top soil for
substantially disturbed             gardening purposes



for all parameters. A study site was rated as heavily dis-           disturbance. The classification of the study sites into this last
turbed when significant differences were detected between            group is the most subjective and is also based on observations
itself and the reference areas (sites 1 and 8) with the use at         in the field (e.g., presence of small pathways, garbage, foot-
least of the parameters Hdune (diversity including coverage)          prints, some ruderals, etc.).
and N (degree of natural diversity). If only one parameter
showed significant differences, this site was classified as only                        RESULTS
moderately disturbed. If no significant changes could be de-
tected, but mean values for both parameters are clearly dif-           Figure 7 shows the MDS plot and dissimilarities between
ferent, the site was classified as having only minor levels of          all 249 plots (B) and two selected sites (A). Even though the




Figure 7. Biplot with multidimensional scaling (MDS) and Bray-Curtis Dissimilarity based on species abundance data with the use of (A) only data
from site 1 (Wolin, undisturbed), which groups on the upper right, and site 11 (Karlshagen, heavily disturbed), which groups in the lower left and (B)
data from all 12 study sites (264 plots and 145 different species). For color version of this figure, see page 1175.




                          Journal of Coastal Research, Vol. 22, No. 5, 2006
                              Tourism and Plant Diversity on Dunes                              1153


                                          plots from each of the study sites mostly group together,
                                          groups of the study sites strongly overlap, and the stress val-
                                          ue is quite high (stress 0.24). However, when only the plots
                                          from reference site 1 (undisturbed) and disturbed site 11 are
                                          being looked at (stress is still high at 0.14), these form dis-
                                          tinct clouds. Major overlapping occurs when all plots are
                                          looked at and the ‘‘gap’’ between the undisturbed and urban
                                          sites is filled. This overlapping is also observed in Figures 8
                                          to 11, which show the box and whisker plots for the old di-
                                          versity index H , the two new diversity parameters Hdune
                                          (quantitative diversity) and N (degree of natural diversity
                                          qualitative diversity) and for the ecological indicator value
                                          (ELLENBERG et al., 1992) for each of the study sites. The num-
                                          ber of plots for each site (n) differs because the width of the
                                          dune area also varies. Generally all parameters show differ-
                                          ences between the sites, and the different types of action
                                          (physical and chemical) are reflected in the results.
                                            The Kruskal-Wallis one-way analysis of variance of ranks
                                          shows significant differences (p 0.01) for all tested diversity
Figure 8. Comparison of tertiary dunes with different levels of recrea-
tional activities with H . Boxes represent the second and third quartiles,
                                          parameters, and a following pairwise multiple comparison
the bar within the boxes indicates the median, and whiskers show the        procedure (Tukey test) revealed significant differences (p
standard deviation. Small dots indicate extreme values (n   number of      0.05) between the site Parking and the remote sites Wolin
plots, m  mean). 1   Wolin, undisturbed; 2 Wolin, near harbor; 3        and Trassenheide for Hdune (Table 3). The analysis for the
Wolin, harbor; 4 Heringsdorf, resort; 5 near Trassenheide; 6 near
Trassenheide; 7   Trassenheide, resort; 8  between Trassenheide and
                                          traditional index H did not reveal as many significant dif-
Karlshagen, undisturbed; 9    Karlshagen, camping; 10   Karlshagen,      ferences as for Hdune. For the degree of natural diversity with-
resort center; 11  Karlshagen, northern resort; 12  parking, north of      in vegetation (N), the Kruskal-Wallis one-way analysis per-
Karlshagen.                                     formed on ranks and a following pairwise multiple compari-
                                          son test (Dunn’s method, Table 3) revealed significant differ-


Table 3. Significant differences according to Kruskal-Wallis one-way analysis performed on ranks and a following pairwise multiple comparison test for
all parameters. Upper right half shows split cells for the diversity parameters H and Hdune . Lower left shows split cells for the level of natural diversity and
the ecological indicator value for nitrogen (significant differences are indicated with ‘‘yes’’).




                            Journal of Coastal Research, Vol. 22, No. 5, 2006
1154                                   Grunewald


                                          dient from the undisturbed sites over sites showing little or
                                          intermediate signs of disturbance toward the heavily dis-
                                          turbed sites seems clear (from the upper left to the lower
                                          right of Figure 7B). This overlapping was also observed in
                                          the box and whisker plots from the different study sites and
                                          parameters in Figures 8 to 11. At undisturbed as well as dis-
                                          turbed sites, plots with high ( 30 species) as well as with
                                          low diversity figures are found. This shows that disturbance
                                          is not evenly distributed on the dunes with respect to inten-
                                          sity, quality (type), or both. In this study, in which patches
                                          of influenced and uninfluenced plots occurred on a small spa-
                                          tial scale together, pooling of the results obtained from at
                                          least three transects per site was still required to allow for a
                                          reliable discrimination between disturbed and undisturbed
                                          sites. Only then were significant differences found between
                                          the study sites and a classification into different ecological
                                          states possible.
                                           The alternative would be to evaluate the whole system and
Figure 9. Comparison of tertiary dunes with different levels of recrea-      document study sites completely with numerous plots—
tional activities with Hdune. Boxes represent the second and third quar-
                                          which is time consuming and impracticable. The common
tiles, the bar within the boxes indicates the median, and whiskers show
the standard deviation. Small dots indicate extreme values (n number        practice, choosing representative sites by the investigator, is
of plots, m  mean). 1  Wolin, undisturbed; 2  Wolin, near harbor; 3     however quite subjective and therefore often critically debat-
  Wolin, harbor; 4  Heringsdorf, resort; 5  near Trassenheide; 6       ed (e.g., BARKMAN, DOING, and SEGAL, 1964; DIERSCHKE,
near Trassenheide; 7 Trassenheide, resort; 8 between Trassenheide
                                          1994). By combining the results of the easily recognizable ‘‘ex-
and Karlshagen, undisturbed; 9    Karlshagen, camping; 10   Karlsh-
agen, resort center; 11   Karlshagen, northern resort; 12   parking,     treme’’ transects—the most influenced and the best pre-
north of Karlshagen.                                served—with those from the subjectively chosen representa-
                                          tive transect, the effect of the subjective element can be min-
                                          imized. As shown in Figures 8 to 10 and Table 3, a good
                                          discrimination between sites influenced by human activities
ences between the rural sites 1 (Wolin), 8 (Trassenheide), and           in a different degree was achieved by this method.
12 (Parking) compared with the urban study sites 10 (Karlsh-             High coverage values and high species numbers at undis-
agen-South), 11 (Karlshagen-North), and the adjacent site 9            turbed sites were observed, whereas disturbed sites exhibited
(Campground). Figure 9 shows box and whisker plots for the             significantly lower mean values, leading to high and low
ecological indicator value (ELLENBERG et al., 1992) for nitro-           Hdune values, respectively (Figure 9). This difference can be
gen. The Kruskal-Wallis one-way analysis (Dunn’s method)              explained by the effect of physical disturbance, which in dune
revealed some, but not all, significant differences between             systems is not homogeneously distributed. Pure physical dis-
study sites, which were also detected with the level of natural
                                          turbance will lead to decreased coverage and a general de-
diversity.
                                          crease in species number, especially in those species that are
 According to the classification system, the sites can be
                                          more sensitive to trampling (e.g., ANDERSEN, 1995; HYL-
grouped into the classes of ecological state shown in Table 4.
                                          GAARD and LIDDLE, 1981; ISERMANN, 1997; LEMAUVIEL and
                                          ROZE, 2003; LIDDLE, 1975; MCDONNELL, 1981).
                                             ´
              DISCUSSION
                                           With respect to eutrophication, most of the nutrient input
 In the study presented here, no clear general trend with             is likely to originate either from fecal pollution or coastal pro-
respect to species richness could be found when individual             tection measures and gardening for aesthetic reasons in the
plots are compared with each other. The MDS with Bray-               vicinity of holiday resorts. Whereas the first effect shows a
Curtis dissimilarity (Figures 7A and 7B) showed a high stress           high degree of patchiness because it is restricted to places
value, and only the presumably undisturbed and obviously              hidden from the beach, the later effect influences large areas
heavily disturbed study sites form distinct clouds. Major             almost homogeneously. Both effects led to increased nutrient
overlapping of all other groups is observed. However, a gra-            availability, resulting in high coverage at plots predominant-


Table 4. Classification of study sites into different classes of ecological state according to different levels of human disturbance.

  Study Sites       Ecological State                           Explanation

1, 2, 8        Undisturbed           No visible signs of permanent human disturbance, sites 1 and 8 were chosen as reference areas
3, 4, 5, 6, 7     Only minor disturbances     Mean values for H dune or N show lower values than for the undisturbed sites
9, 12         Medium disturbances       Either H dune or N show significant differences compared with one of the undisturbed sites
10, 11         Heavy disturbances       Both H dune and N show significant differences compared with both undisturbed sites
None in this study   Irreversible disturbances    Remobilization of the dune and/or large areas of intensive technical coastal defense measures




                           Journal of Coastal Research, Vol. 22, No. 5, 2006
                             Tourism and Plant Diversity on Dunes                             1155




Figure 10. Comparison of tertiary dunes with different levels of recre-     Figure 11. Comparison of tertiary dunes with different levels of recre-
ational activities with the use of the degree of natural diversity (typical   ational activities with the use of the ecological indicator value (unweight-
dune diversity). Boxes represent the second and third quartiles, the bar     ed mean, Ellenberg et al., 1992) for nitrogen. Boxes represent the second
within the boxes indicates the median, and whiskers show the standard      and third quartiles, the bar within the boxes indicates the median, and
deviation. Small dots indicate extreme values (n number of species, m      whiskers show the standard deviation. Small dots indicate extreme val-
  mean). 1   Wolin, undisturbed; 2   Wolin, near harbor; 3   Wolin,    ues (n  number of plots, m   mean values). 1 Wolin, undisturbed; 2
harbor; 4  Heringsdorf, resort; 5  near Trassenheide; 6  near Tras-      Wolin, near harbor; 3   Wolin, harbor; 4   Heringsdorf, resort; 5
senheide; 7   Trassenheide, resort; 8   between Trassenheide and      near Trassenheide; 6   near Trassenheide; 7    Trassenheide, resort; 8
Karlshagen, undisturbed; 9    Karlshagen, camping; 10   Karlshagen,      between Trassenheide and Karlshagen, undisturbed; 9 Karlshagen,
resort center; 11  Karlshagen, northern resort; 12   parking, north of    camping; 10    Karlshagen, resort center; 11    Karlshagen, northern
Karlshagen.                                   resort; 12  parking, north of Karlshagen.



ly influenced by nutrient input. The calculated ecological in-          are made up of well-sorted aeolian sand deposits, dunes are
dicator values (ELLENBERG et al., 1992) clearly indicate sig-          considered very fragile ecosystems and very sensitive to
nificant differences in nutrient availability between the sites          trampling. Any further frequent mechanical damage is likely
as shown in Figure 11 and Table 3. However, these values do           to exceed the level of intermediate disturbances, leading to
not completely support the hypothesis that eutrophication is           decreasing species numbers, but this remains a hypothesis at
the major factor leading to qualitative changes in species di-          this stage.
versity. The qualitative changes in species diversity reflect            With the use of the traditional index H , fewer and also
the changing habitat conditions and the index N (degree of            different significant differences were detected (Table 3) com-
natural diversity) clearly showed differences between the            pared with Hdune. Because even the most obvious differences
sites, but these changes are most likely complemented by ef-           between the mostly undisturbed reference site 1 on Wolin
fects from the larger species pools at the urban sites (3, 10,          and the heavily disturbed site 11 (holiday resort of Karlsh-
and 11). Neighboring effects from nearby gardens, parks, or           agen) with low coverage values (Figure 4) were not detected,
ruderal sites might play an important role for the species            these results support the use of Hdune as a better index of
composition on the dunes (species pool hypothesis; PARTEL et
                            ¨             disturbance. With respect to coverage, the common procedure
al., 1996).                                   of calculating H regularly neglects the trampling effect on
  The special features of the naturally nutrient-limited dune          plant cover because of normalization to 100%. As shown in
ecosystem lead to an ambivalent reaction to anthropogenic            Figure 6 and in GRUNEWALD and SCHUBERT (2005), H is
activity consisting of a combination of physical and chemical          consequently not able to detect such a combination of distur-
disturbances that often overlap. This system is complicated           bances. In fact, with respect to eutrophication and neighbor-
by the level and frequency of human disturbance. The inter-           ing effects, the use of H for calculating the degree of natural
mediate disturbance hypothesis (CONNELL, 1978) shows that            diversity as described above sometimes will result in N val-
low levels of disturbances that occur frequently, creating hab-         ues well above 1. If, e.g., the three dominant species in C2
itats for some species, will lead to a habitat patchiness pat-          (Figure 6) happen to be alien species and would be omitted
tern with higher diversity than the undisturbed original hab-          from calculation, H would actually rise (from C2 to C3, Fig-
itat. Applying this hypothesis on the special conditions of           ure 6). This makes an interpretation of the degree of natural
dune habitats, the following can be concluded. Grey dune eco-          diversity with H much more complicated. If normalization of
systems are characterized by a high level of natural (me-            total coverage is left out, the newly proposed Hdune becomes
chanical) disturbances and normally already show a patch-            a fair measure of disturbance status.
work of different stages of dunal succession. Because dunes            However, in many cases, plots from disturbed and undis-



                           Journal of Coastal Research, Vol. 22, No. 5, 2006
1156                               Grunewald


turbed sites still overlap. Plots with a lower degree of natural     plays a major role for the condition of the dunes along the
diversity at undisturbed sites can be explained by the occa-       Pomeranian Bight. Remote sites 1 and 8 showed no signs of
sional presence of alien species on all dunes (in fact, in almost     permanent disturbance and were therefore chosen as exam-
all central European habitats). This results from both the        ples for the undisturbed ecological state. Tourist numbers
general botanical pollution as well as eutrophication effects       were also lower at these sites, and KAMMLER (2003) found
by atmospheric deposition of nitrogen that was estimated at        the highest numbers of visitors on beaches of the southern
an average of 50 kg ha 1 yr 1 in central Europe (ELLENBERG,        Baltic Sea coast in the immediate vicinity of holiday resorts
1996). In this study, C. canadensis was found on some plots        and parking lots. Hence, low or little access seems to be the
on Wolin Island and C. introflexus was occasionally found on        best conservation method because most people are reluctant
Usedom, as well as on Wolin. Campylopos introflexus is con-        to walk any distance from parking areas or hotels to the
sidered an invasive species in more western parts of central       beach. This effect seems to be overlaid by the level of control
Europe (e.g., on North Sea dunes in Germany or The Neth-         of the beach (and the dunes). Sites that were adjacent to life-
erlands; PETERS and POTT, 1999), but further east it has not       guard posts or that were easily observed from the beachside
spread as much. This could be because of climatic reasons or       promenade, official dune crossings, or the beach showed less
because of less time to establish itself so far in the Baltic Sea     change than others (sites 4–7). If the promenade is on the
region.                                  higher dunes overlooking the younger ones, trampling might
  Overall diversity according to Hdune seems to be an appro-       be low, but the promenade itself will destroy a large portion
priate measure for quantitative changes on coastal dunes,         of the dune habitat. If, however, the promenade lies some
and the level of natural diversity (N) could serve as an ad-       distance inland and the dunes cannot be observed, then the
ditional independent measure for qualitative changes for de-       overall level of trampling on the dunes is higher (e.g., north-
termining the strength of the anthropogenic disturbance. It        ern dunes in Karlshagen at site 11). A possible solution could
is capable of detecting eutrophication and any neighboring        be the installation of a few well-maintained, elevated look-
effects, whereas physical effects are already reflected by         outs that protrude from the promenade (further inland) onto
Hdune. The pooling of data from the so-called extreme tran-        the dunes and that would probably become popular for visi-
sects with those from a representative transect diminishes        tors.
the remaining element of subjectivity.
                                                 ACKNOWLEDGMENTS
CONCLUSIONS AND MANAGEMENT PROPOSALS
                                       This work was supervised by Prof. H. Schubert, University
  The method presented can be seen as a new approach to-         of Rostock. Research in Poland was conducted in cooperation
ward analyzing human disturbance on dunes. The unified           with MGR Tomasz Łabuz from the University of Szczecin.
plot size and the use of the relative coverage percentage (with
respect to the plot size) was chosen so that the adapted index                  LITERATURE CITED
Hdune incorporates not only information on the relative abun-
                                     ANDERSEN, U.V., 1995. Resistance of Danish coastal vegetation
dance of species and species richness but, in contrast with         types to human trampling. Biological Conservation, 71, 223–230.
H , also includes information on plant cover or species den-       BARKMAN, J.J.; DOING, H., and SEGAL, S., 1964. Kritische Bemer-
sity (as an additional component of diversity and dune sta-         kungen und Vorschlage zur Quantitativen Vegetationsanalyse.
                                                  ¨
bility that is being reduced by mechanical damages). Tram-         Acta Botanica Neerlandica, 13, 394–419.
                                     BAUR, B. and EHRHARDT, A., 1995. Habitat fragmentation and hab-
pling will cause heavy damage along clearly defined path-          itat alterations: principal threats to most animal and plant spe-
ways, but adjacent areas might not be affected at all, so the        cies. Gaia, 4, 221–226.
complete range of different levels of disturbance needs to be       BOGAERT, G. and LEMEUR, R., 1995. The Use of Diversity Indices in
documented. In the vicinity of holiday resorts, dune species        the Detection of Succession and Ecological Disturbances in Bel-
                                      gian Dune Vegetations (at Koksijde). Mededelingen Faculteit Land-
composition clearly shifts toward more common ruderal gen-
                                      bouwkundige en Toegepaste Biologische Wetenschappen Universiteit
eralists, which could be jointly linked to increasing nutrient       Gent 60(1), 61–68.
levels on those dunes or the neighboring effects from a nearby      BRAUN-BLANQUET, J., 1964. Pflanzensoziologie. Grundzuge der Ve-
                                                                  ¨
holiday resort. In addition, dunes have often also been plant-       getationskunde, 3rd edition. Berlin: Springer, 865p.
ed with a variety of different species. Some are useful in re-      BUNDESAMT FUR NATURSCHUTZ (eds.), 1997. Biodiversitat und
                                              ¨                      ¨
                                      Tourismus. Berlin: Springer Verlag, 339p.
ducing erosion on degraded dunes (R. rugosa, Elaeagnus sp.),       CONNELL, J.H., 1978. Diversity in tropical rainforests and coral
but others were introduced purely for aesthetic reasons (e.g.,       reefs. Science, 199, 1302–1310.
S. albus); furthermore, most of these are alien species that       CORNELL, H.V., 1999. Unsaturation and regional influences on spe-
can oust typical native dune species, reducing natural biodi-        cies richness in ecological communities: a review of the evidence.
                                      Ecoscience, 6(3), 303–315.
versity (ISERMANN, 1997). Thus, as was shown in this study,
                                     DIERSCHKE, H., 1994. Pflanzensoziologie: Grundlagen und Method-
the degree of natural diversity of the dunes can decrease          en. Stuttgart: Ulmer Verlag, 683p.
while biodiversity indices detect a rise in quantitative diver-      ELLENBERG, H., 1996. Vegetation Mitteleuropas mit den Alpen in
sity, which makes it necessary to look at the qualitative          okologischer, dynamischer und historischer Sicht. 5. Aufl. Stutt-
                                      ¨
changes on a species level.                         gart: Ulmer Verlag, pp. 511–561.
                                     ELLENBERG, H.; WEBER, H.E.; DULL, R.; WIRTH, V.; WERNER, W.,
                                                       ¨
  Because study sites were inter alia selected according to        and PAULISSEN, D., 1992. Zeigerwerte von Pflanzen in Mitteleu-
their accessibility, the results suggest that access to the         ropa. Scripta Geobotanica, 18, 1–258. [Includes an English sum-
beach, as well as the development of recreational facilities,        mary]




                       Journal of Coastal Research, Vol. 22, No. 5, 2006
                          Tourism and Plant Diversity on Dunes                          1157


GRIME, J.P., 2002. Plant Strategies, Vegetation Processes, and Eco-    LANDESAMT FUR FORSTEN UND GROSSSCHUTZGEBIETE MECKLEN-
                                              ¨
 system Properties, 2nd edition. Chichester, New York: J. Wiley &      BURG-VORPOMMERN and NATIONALPARKAMT VORPOMMERSCHE
 Sons. 417p.                                BODDENLANDSCHAFT (eds.), 2002. Nationalparkplan Vorpommer-
GRUNEWALD, R., 2004. Biodiversitat und Tourismus: Auswirkungen
                  ¨                    sche Boddenlandschaft. Malchin, Germany.
 des Fremdenverkehrs auf die Strand- und Dunenvegetation der
                          ¨            LEMAUVIEL, S. and ROZE, F., 2003. Response of three plant com-
                                                   ´
 Pommerschen Bucht. In: BUNDESAMT FUR NATURSCHUTZ (ed.),
                       ¨               munities to trampling in a sand dune system in Brittany (France).
 Treffpunkt Biologische Vielfalt 4. Berlin: Springer-Verlag.        Environmental Management, 31(2), 227–235.
GRUNEWALD, R. and SCHUBERT, H., 2005. The definition of a new        LIDDLE, M.J., 1975. A theoretical relationship between the primary
 plant diversity index ‘‘Hdune ’’ for assessing human damage on       production of vegetation and its ability to tolerate trampling. Bi-
 coastal dunes—derived from the Shannon index of entropy—H .        ological Conservation, 8, 251–255.
 Ecological Indicators, 199, 1–21.                    LIDDLE, M.J. and GREIGH-SMITH, P., 1975. A survey of tracks and
HAEUPLER, H., 1982. Evenness als Ausdruck der Vielfalt in der Veg-      paths in a sand dune ecosystem. I. Soils. Journal of Applied Ecol-
 etation. Vaduz, Liechtenstein: J. Cramer, 260p.              ogy, 12, 893–908.
HAEUPLER, H. and MUER, T., 2000. Bildatlas der Farn- und Blue-       MAGURRAN, A.E., 1988. Ecological Diversity and its Measurements.
 tenpflanzen Deutschlands. Stuttgart: Ulmer Verlag, 759p.          Princeton, New Jersey: Princeton University Press, 179 p.
HELLEMAA, P., 1998. The development of coastal dunes and their       MCDONNELL, M.J., 1981. Trampling effects on coastal dune vege-
 vegetation in Finland. Fennia, 176(1), 111–221.              tation in the Parker River National Wildlife Refuge, Massachu-
HESP, P.A., 1991. Ecological processes and plant adaptations on       setts, USA. Biological Conservation, 21, 289–301.
 coastal dunes. Journal of Arid Environments, 21, 165–191.        MULLER, D., 2004. Naturraumliche Grundlagen Mecklenburg-Vor-
                                       ¨            ¨
HILL, M.O.; ROY, D.B., and THOMPSON, K., 2002. Hemeroby, urban-       pommern. In: BERG, C.; DENGLER, J.; ABDANK, A., and ISER-
 ity and ruderality: bioindicators of disturbance and human impact.     MANN, M. (eds.), Die Pflanzengesellschaften Mecklenburg-Vorpom-
 Journal of Applied Ecology, 39, 708–720.                  merns und ihre Gefahrdung. Jena, Germany: Weissdorn-Verlag,
                                                 ¨
HOBOHM, C. and PETERSEN, J., 1999. Zur Artenvielfalt von Zwerg-       pp. 22–26.
 binsengesellschaften. Mitteilungen Bad. Landesver. Naturkunde      PARTEL, M.; ZOBEL, M.; ZOBEL, K., and VAN DER MAAREL, E., 1996.
                                      ¨
 und Naturschutz, N.F., 17, 313–318.                    The species pool and its relation to species richness: evidence from
HURTIG, T., 1957. Physische Geographie von Mecklenburg. Greifs-       Estonian plant communities. Oikos, 75, 111–117.
 wald. Berlin: VEB Deutscher Verlag der Wissenschaften, 252p.       PETERS, M. and POTT, R., 1999. Natur und Tourismus auf Norder-
HYLGAARD, T. and LIDDLE, M.J., 1981. The effect of human tram-        ney. Abhandlungen aus dem Westfalischen Museum fur Naturkun-
                                                         ¨          ¨
 pling on a sand dune ecosystem dominated by Empetrum nigrum.        de, 61 (supplement) 174S.
 Journal of Applied Ecology, 1981, 559–569.                PICKETT, S.T.A. and WHITE, P.S. (eds.), 1985. The Ecology of Natural
ISERMANN, M., 1995. Dunes in contradiction with different interests.     Disturbance and Patch Dynamics. Orlando, Florida: Academic
 An example: the camping-ground prerow (Darß/Baltic Sea). In:        Press, 472p.
 SALMAN, A.H.P.M., and BONAZOUNTAS, M. (eds.), Coastal Man-        PIOTROWSKA, H., 1979. Natural resistance against mechanical de-
 agement and Habitat Conservation. Marathon, Greece; Proceedings      struction in plant communities on Baltic coast dunes. Biologia,
 of the 4th EUCC Conference. Publ. EUCC, Volume I, pp. 439–449.       1979(1), 5–13.
ISERMANN, M., 1997, Vegetations- und standortskundliche Unter-       PIOTROWSKA, H., 1989. Natural and anthropogenic changes in sand-
 suchungen in Kustendunen Vorpommerns. Doctoral thesis. Greifs-
          ¨    ¨                         dunes and their vegetation on the southern Baltic coast. Perspec-
 wald, Germany: University of Greifswald, 323p.               tives in Coastal Dune Management, 1989, 33–40.
ISERMANN, M., 2001. Kornungsverhaltnisse in Kustendunen Meck-
             ¨       ¨      ¨  ¨         PIOTROWSKA, H. and STASIAK, J., 1982. Naturalne i antropogeniczne
 lenburg-Vorpommerns. Proceedings, Forschung und Naturschutz        zmiany strefore flory nacziniowej bezles  ´nych wydm nadmorskich
 in Sandlebensraumen (Freiburg, Germany), pp. 61–64.
         ¨                             Mierrzei Wislanej [The natural and anthropogenic zonal changes
                                             ´
JANKE, W., 1971. Beitrag zur Entstehung und Alter der Dunen der
                               ¨        of vascular flora on the non-forest coastal dunes of the Vistula
 Lubminer Heide soeie Peenemunde-Zinnovitzter Seesandebene.
                  ¨                    Spit]. Fragmenta Floristica et Geobotanica, 28(2), 371–396.
 Greifswald. Wissenschaftliche Zeitschrift Ernst-Moritz-Arndt Univ-    REINHARD, H., 1951. Karten zur Klimabewertung bei der Boden-
 ersitat, Math.-Nat. Wiss. R., 20(1/2), 39–54.
    ¨                                   schatzung in Mecklenburg. Angewandte Meteorologie, 1(1).
                                         ¨
JESCHKE, L., 1985. Vegetationsveranderungen in den Kustenland-
                   ¨          ¨        REINHARD, H., 1962. Klimatologie. VEB—Topographischer Dienst
 schaften durch Massentourismus und Nutzungsintensivierung.         Schwerin, I. Natur des Landes. Atlas f. d. Bezirke Rostock, Schwe-
 Archiv Naturschutz und Landschaftsforschung, 25, 223–236.         rin and Neubrandenburg.
KAMMLER, M., 2003. Intensitat und raumliche Struktur des Tour-
                ¨     ¨                RODGERS, J.C., 2002. Effects of human disturbance on the dune veg-
 ismus in der Kustenregion Warnemunde—Kuhlungsborn. Kiel,
           ¨           ¨    ¨            etation of the Georgia Sea Islands. Physical Geography, 23(1), 79–
 Germany: University of Kiel, Diploma thesis, 121p.             94.
KLIEWE, H. and RAST, H., 1979. Geomorphologische und mikrom-        RODGERS, J.C. and PARKER, K.C., 2003. Distribution of alien plant
 agnetische Untersuchungen zu Habitus, Struktur und Genese des       species in relation to human disturbance on the Georgia Sea Is-
 Zinnovitzer-Trassenheider Strandwallsystems und seiner Du-   ¨     lands. Diversity and Distributions, 9(5), 385–398.
 nen.—Gotha Leipzig. Petermanns Geographische Mitteilungen,        SEIDEL, W., 2001. Tourismus in Mecklenburg-Vorpommern 1999.
 123(4), 225–242.                              Statistische Sonderhefte des Statistischen Landesamtes Mecklen-
KNAPP, H.D., 1996. Belastung und Schutz der Boddenlandschaften        burg-Vorpommern Jg. 11, H. 3. 59p.
 an der deutschen Ostseekuste. In: BUCHWALD, K., and ENGEL-
               ¨                      SHANNON, C.E. and WEAVER, W., 1949. A Mathematical Theory of
 HARDT, W. (eds.), Umweltschutz—Grundlagen und Praxis, Band 6/       Communication. Chicago: University of Illinois Press. 125p.
 II. Bonn: Economica Verlag, pp. 76–112.                 VAN DER MAAREL, E., 1971. Plant species diversity in relation to
KUTIEL, P.; ZHEVELEV, H., and HARRISON, R., 1999. The effect of       management. In: DUFFEY, E., and WATT, A.S. (eds.), The Scientific
 recreational impacts on soil and vegetation of stabilised coastal     Management of Animal and Plant Communities for Conservation.
 dunes in the Sharon Park, Israel. Ocean and Coastal Management,      Oxford, UK: Blackwell Scientific Publications, pp. 45–63.
 42(12), 1041–1060.                            VON NORDHEIM, H. and BOEDEKER, D. (eds.), 1998. Red list of bio-
ŁABUZ, T. and GRUNEWALD, R., 2007. Studies on vegetation cover        topes and biotope complexes of the Baltic Sea, Belt Sea and Kat-
                 ´
 of the youngest dunes of the Swina Gate Barrier (western Polish      tegat. Baltic Sea Environment Proceedings (Helcom). No. 75,
 coast). Journal of Coastal Research. In press.               115 p.




                        Journal of Coastal Research, Vol. 22, No. 5, 2006