Andersen 1995
Biological Conservation71 (1995) 223 230
,~ 1995 Elsevier Science Limited
Printed in Great Britain. All rights reserved
0006-3207/'95/$09.50+.00
ELSEVIER 0006-3207(94)00031-X
RESISTANCE OF DANISH COASTAL VEGETATION TYPES TO
H U M A N TRAMPLING
Ulla Vogt Andersen
Botanical Section, Department of Botany, Dendrology and Forest Genetics, Royal Veterinary & Agricultural University,
Rolighedsvej 21, DK-1958 Frederiksberg C, Denmark
(Received 14 January 1994; revised version received 20 May 1994; accepted 8 June 1994)
Abstract 1970; Hosier & Eaton, 1980), and loss of biodiversity
Five coastal communities under influence o f human tram- (McDonell, 1981). Because the impact results in soil
pling, i.e. a natural salt marsh, a natural dune, a man- modification, the changes in vegetation composition
made dune, and two man-made coastal grasslands, were and structure can be irreversible (Beeftink, 1979). Low
studied. The vegetation of five paths (one in each com- levels of trampling can, however, have a beneficial infl-
munity) created and sustained by, human trampling was uence on species diversity (Liddle & Greig-Smith, 1975b;
analysed with respect to floristic composition and species Boorman & Fuller, 1977), by keeping the communities
cover. The total number of vascular plant species, species in a dynamic stage (Magnusson, 1986).
diversity', and the total vegetation cover were significantly Careful planning of recreational impact is necessary
reduced. The therophytes and hemicryptophytes were sig- and a knowledge of the resistance of seashore communi-
nificantly reduced, while the geophytes were indifferent to ties is an important basis for planners. The aims of the
the impact o f human trampling. Man-made and natural p, esent study were to compare the vulnerability of five
dunes were most vulnerable, coastal grasslands were different plant communities, i.e. a natural salt marsh, a
intermediate, and the salt marsh was the most resistant natural dune, a man-made dune, a man-made coastal
to human trampling. grassland under management, and a man-made coastal
grassland without management. These communities were
Keywords: human trampling, disturbances, species re- analysed with emphasis on (1) the trampling-induced
sistance, man-made ecosystems, coastal communities. changes in vegetation composition and species diversity;
(2) the ability of individual species to withstand injury; and
(3) the possible differences in resistance between man-
INTRODUCTION
made and natural coastal areas and between managed
For holiday makers of Northern Europe seashores rank and unmanaged plant communities. Plant nomenclature
among the most attractive areas. At present, employ- follows Hansen (1981).
ment and income in many coastal areas are largely
dependent on the tourist industry, and tourism is a growth
STUDY AREAS
sector with a large potential in economic terms (Meijer,
1992; Jensen, 1993). Tourist activities, however, can The studies were made in Denmark in Koge Bay Sea-
have severe effects on coastal areas, and contribute to side Park and Olsemagle Revle (Fig. 1), both offshore
their destruction and reduction of their nature and barriers situated 20 km apart, on the coast of the Baltic
recreational values. A direct result of leisure activities is Sea. Olsemagle Revle was created by natural processes:
disturbance to the flora and fauna. Several researchers interaction of sand deposition and plant growth during
have studied the impact of human trampling in dunes the period 1900-1930. In contrast, Koge Bay Seaside
and other seashore plant communities. The natural re- Park was planned by architects and constructed for
sponses are soil compaction (Bates, 1935; Liddle & recreational purposes, in the years 1978-1980.
Greig-Smith, 1975a), reduction in soil organic matter Olsemagle Revle covers 200 ha and consists of a 5
(Boorman & Fuller, 1977; Hylgaard & Liddle, 1981), km sandy beach with a well-developed strandline vege-
decrease of vegetation cover (Burden & Randerson, tation, a row of dunes with a height of approximately 3
1972; Bowles & Maun, 1982), decrease in biomass pro- m, and a salt marsh with a Phragmites-Scirpus swamp
duction (Edmond, 1962; Liddle & Greig-Smith, 1975b), on the landward side of the barrier which encloses a
reduction in number of flowering species (Goldsmith et lagoon. The salt marsh has never been grazed or mown.
al., 1970; Hylgaard, 1980), disappearance of vulnerable The only man-made facility is a constructed dam with
species (Bates, 1935; Chappel et al., 1971), creation of an access road. A comprehensive study of the vegeta-
paths (Bayfield, 1973; Hylgaard & Liddle, 1981), tion was made by Gravesen and Vestergaard (1969).
erosion (Frederiksen, 1977; Carlson & Godfrey, 1989), Koge Bay Seaside Park covers 500 ha and consists
interference in the natural succession (Goldsmith et al., of 8 km of sandy beach with 3-m high artificial dunes
223
224 Ulla Vogt Andersen
1 Kege Ba' _Seaside Park
N
,,, .., t.~....o,~. .
)a~'~'~'~3~ ~ ~ Lagoon
¢ • f~.~ t';,..... ~\,.:--,~-~.,j....~,...<
~ :~ .. ~~ ,.,...~..~. ":
~,...
" ~ ' ~ Baltic Sea
LEGEND:
L~] Parking lot
[] Dunes
i~! Mainlandbehind original coastline
Road with vehicle access
Constructed path/bike lane
. . . . Path created by human trampling
Sites where surveys a r e undertaken
Study area
Fig. 1. Map showing the location of the study areas in Denmark (inset bottom left), and the position of the study sites 1-5 in
Kege Bay Seaside Park (top) and Olsemagle Revle (bottom right).
constructed on two long, narrow offshore barrier the artificial dunes Ammophila arenaria was planted in
islands. The new artificially advanced coastline is con- 1978. All other species occurring in the area in 1992
nected with the original natural coastline by bridges have been invading as a result of primary succession
and dykes. Behind the offshore barriers are six lagoons. (Hansen & Vestergaard, 1986). Several facilities for visitors
The following vegetation types and plant communities are are available: harbours, roads, parking areas, toilets, life
present: sandy beach, sand dunes, grasslands, former guards, ice-cream stands, etc. Being closer to Copen-
tidal meadows, lagoon shores, roadsides, shrubberies, and hagen city and having all the recreational facilities,
small planted groves. The grasslands were sown in 1979 Kege Bay Seaside Park is visited by more than 500,000
with a seed mixture of grasses consisting of Festuca rubra, people each year. In contrast, Olsemagle Revle receives
F. arundinaeea, Lolium multiflorum and L. perenne. In less than 100,000 visitors in the same period.
T r a m p l i n g o f Danish coastal vegetation 225
Table 1. Percentage frequency of 32 taxa including bare soil in +, trampled path centre, and -, untrampled surroundings at five sites
Site 1, m o w n grassland; site 2 u n m o w n grassland; site 3, artificial dune; site 4, natural salt marsh; site 5, natural dune, The species
are tested by likelihood-ratio X 2 for whether their occurrence is independent of trampling impact. Results are divided into groups ac-
cording to their response. G 2 -- Likelihood ratio value (d.f. -- 1) *** p < 0.001, **p < 0.01, * p < 0.05. Favoured, species occurring
only o n the paths; Vulnerable, species occurring only in untrampled situations; Indifferent, species occurring in b o t h situations.
Species Site 1 Site 2 Site 3 Site 4 Site 5 G2 Significance
+ + + + - +
Favoured
Bare soil 100 100 100 25 100 100 85 - 100 60 52.099 ***
Elytrigia repens -- -- -- 90 30 -- 30 -- 3-897 *
Festuca arundinacea 10 -- 80 5 . . . . 20.103 ***
Glaux maritima -- -- -- 55 15 -- -- 4.916 *
Leymus arenarius . . . . . . . . 75 25 5.788 *
Vulnerable
Bryophytes 70 90 65 70 -- 100 -- 20 60 85 23-971 ***
Lichens -- -- -- 5 80 -- -- 10 28.338 ***
Cirsium arvense 5 30 15 15 55 -- -- -- 13-101 ***
Cerastium semidecandrum . . . . 30 -- -- 25 15.890 ***
Erigeron acer 5 -- 5 5 5 55 -- -- -- 6.220 *
Holcus lanatus -- -- 95 . . . . 28.338 ***
Hypochoeris radicata . . . . . . . . 15 60 6-220 *
R u m e x acetosella 10 5 95 -- -- 20 50 25.034 ***
Tanacetum vulgare . . . . . . 15 80 10.700 **
Trifolium arvense 10 90 65 -- 20 I __ -- 30.969 ***
Viola tricolor . . . . . 5 65 12.978 ***
Indifferent
Agrostis stolonifera -- -- 90 90 -- -- 100 95 -- -- 0.021 ns
Ammophila arenaria -- -- 90 100 . . . . . 0.130 ns
Ammophila x Calamagrostis . . . . . 5 95 100 0-000 ns
Chamaenerion angustifolium . . . . 30 65 . . . . 2.911 ns
Dactylis glomerata 15 30 65 60 . . . . 0 142 ns
Festuca rubra I00 100 100 100 30 100 100 -- 20 0.088 ns
Festuca ovina -- -- -- 80 100 . . . . 0.543 ns
Hippopha# rhamnoides 30 -- 5 30 . . . . . 0.082 ns
Honckenya peploides . . . . 25 30 0.096 ns
Lathyrusjaponicus -- 60 45 -- -- 5 30 0.000 ns
Odontites verna . . . . . . 95 95 -- 0.00 ns
Phragmites australis 25 . . . . 55 60 -- 0.666 ns
Plantago maritima . . . . . 100 80 -- I 0-543 ns
Poa pratensis -- 10 10 15 -- 5 5 15 25 1.629 ns
Scirpus maritimus . . . . . 45 65 -- -- 0.821 ns
Taraxacum spp. 100 100 95 95 95 100 -- -- 10 10 0-021 ns
Trifolium repens 30 5 100 100 -- 5 5 -- 0-677 ns
METHODS c e n t r e o f all five p a t h s 0.6 m a p a r t , a n d a s i m i l a r set o f
q u a d r a t s p l a c e d i n t h e v e g e t a t i o n a l o n g s i d e , 2.5 m f r o m
Five paths, with a width of at least 1 m, and their sur- the centre of the path. Other studies have shown that
r o u n d i n g c o m m u n i t i e s w e r e s e l e c t e d as s u r v e y sites. people tend to follow paths very strictly (Bayfield,
Three are situated in Koge Bay Seaside Park: one on 1973; F r e d e r i k s e n , 1977; K a r d e l l , 1978; H y l g a a r d ,
t o p o f t h e a r t i f i c i a l d u n e s (site 3) a n d t w o i n t h e g r a s s - 1980), a n d t h e r e f o r e t h e v e g e t a t i o n o u t s i d e t h e p a t h s
l a n d s b e h i n d t h e d u n e s . T h e g r a s s l a n d a t J~egerso (site c a n b e r e g a r d e d as u n t r a m p l e d . A t o t a l o f 100 u n t r a m -
1) is c u t o n c e a y e a r a n d H i p p o p h a ~ r h a m n o i d e s is c o n - p i e d a n d 100 t r a m p l e d p l o t s w e r e t h u s a n a l y s e d .
t r o l l e d w i t h h e r b i c i d e s , w h i l e t h e g r a s s l a n d a t Lille In each plot the cover of individual species was mea-
Vejleso (site 2) is u n d e r g o i n g n a t u r a l d e v e l o p m e n t w i t h o u t sured using the Hult-Sernander-DuRietz scale of cover
any management. Two paths are situated on Olsemagle c l a s s e s ( M a i m e r , 1974). I n t h e s t a t i s t i c a l t r e a t m e n t o f
R e v l e : o n e o n t h e d u n e t o p (site 5) a n d o n e i n t h e s a l t data the cover values were transformed into the follow-
m a r s h (site 4). ( F i g . 1). T h e p a t h s i n K o g e B a y S e a s i d e i n g p e r c e n t a g e v a l u e s ( H a n s e n & J e n s e n , 1972): less
P a r k a r e u s e d b y a p p r o x i m a t e l y 10 v i s i t o r s e a c h d a y t h a n 1/16 = 2 % ; 1 / 1 6 - 1 / 8 = 9%; 1 / 8 - 1 / 4 = 18%; 1 / 4 - 1 / 2
(yearly average, according to the Park headquarters). = 36%; a n d m o r e t h a n 1/2 = 7 2 % . F r e q u e n c y o f species
T h e u s e o f t h e p a t h s a t t 0 1 s e m a g l e R e v l e is n o t k n o w n . w a s m e a s u r e d as t h e p e r c e n t a g e o c c u r r e n c e in t h e 1 - m 2
H o w e v e r , f r o m c o u n t s d u r i n g t h e s u r v e y it is e s t i m a t e d plots. Species diversity was measured as number of
t h a t t h e p a t h s a r e u s e d b y five v i s i t o r s d a i l y . species per plot based on an average from 20 plots.
I n J u l y 1993, 20 1 - m 2 q u a d r a t s w e r e p l a c e d a l o n g t h e P l a n t n o m e n c l a t u r e f o l l o w s H a n s e n (1981).
226 Ulla Vogt Andersen
Table 2. Percentage frequencies of 33 taxa grouped according to their relative occurrence on either trampled path (+) or untrampled
surroundings (-)
For explanation of sites and categories, see Table 1.
Species Site 1 Site 2 Site 3 Site 4 Site 5
+ - + + -- + -- +
Favoured
Lolium perenne -- 20 -- -- 5 -- -- --
Poa annua -- -- 5 . . . . . . .
Spergularia marina . . . . . 35 --
Tussilago fitrfara 5 . . . . . . .
Vulnerable
Aira praecox . . . . . . . 10
Anthyllis vulneraria -- 5 -- 5 . . . . .
Arenaria serpyllifolia . . . . 5 -- --
A triplex prostrata ssp. prostrata . . . . . . 5
Carduus crispus . . . . 5 -- -- --
Conyza canadensis -- -- -- 5 -- -- 5
Galium verum . . . . 5 -- -- --
Leontodon autumnalis . . . . . . . 10
Melilotus alba -- 5 -- 5 . . . . .
Plantago lanceolata -- --- 15 . . . .
Rosa canina -- -- 5 . . . . . .
Rumex crispus . . . . . 5 -- -- --
Tragopogon pratensis -- -- -- 10 . . . .
Trifolium dubium -- -- 30 -- --
Triglochin maritimum . . . . . . 5
Senecio vernalis . . . . 10 -- -- 20 85
Vicia sativa ssp. angusti~blia -- 10 -- 5 . . . . .
Vicia hirsuta -- -- 35 . . . . . .
Indifferent
Artemisia vulgaris 10 15 5 . . . . .
Carex arenaria . . . . 30 5 -- -- --
Cerastium fontanum ssp. triviale 5 5 -- 5 . . . . .
Elytrigia juneeiforme . . . . . 35 10
Juncus gerardi . . . . . . 15 15
Lotus tenuis . . . . . 10 25 -- --
Potentilla anserina . . . . 10 15
Rosa rugosa -- -- 5 10 30 -- --
TriJolium pratense 15 5 -- 10 . . . .
Triglochin palustre . . . . 5 15 -- --
Vicia craeca . . . . . 5 10
Species resistance to h u m a n trampling and their abil- all sites is shown in Tables 1 and 2. Figure 2 shows the
ity to regenerate is to a large extent dependent u p o n n u m b e r o f species in each plant community. The high-
the position o f their growth points and surviving buds est n u m b e r o f species (24), was f o u n d in the u n m o w n
(Burden & Randerson, 1972; Liddle & Greig-Smith, grassland o f site 2. The total n u m b e r o f species was
1975b; H y l g a a r d & Liddle, 1981). Raunki~er's (1934) reduced significantly by the trampling impact at all
life-form classification o f species is therefore used here. sites (Table 3), although in the m o w n grassland (site 1)
G r a m i n o i d s (including Poaceae, Juncaceae and Cyper- the same n u m b e r occurred in b o t h situations, and in
aceae) are treated as a special g r o u p because o f their the natural salt m a r s h (site 4) little difference was
flexible, linear leaves, and their ability to regenerate found. The reduction was m o s t noticeable a m o n g the
after trampling (Bates, 1935; Chappell e t a L , 1971; forbs, in contrast to the grasses o f which there are
Page e t a l . , 1985). generally m o r e species on the trampled path.
Paths and u n t r a m p l e d surroundings were c o m p a r e d The species diversity o f the communities was gener-
with respect to the frequency o f occurrence o f 30 ally low and the highest diversity (>8 species/m 2) was
species, together with bare soil, lichens, and bryo- f o u n d in the u n m o w n , u n t r a m p l e d grassland (site 2)
phytes, using a likelihood-ratio X2-test. The remaining (Fig. 3). The diversity decreased significantly as a result
n u m b e r o f species (22) occurred too infrequently for o f the trampling impact. The reduction was largest in
the X2-test to be valid. the natural dune and in the u n m o w n grassland,
whereas no reduction occurred on the path in the salt
marsh, site 4.
RESULTS
The total vegetation cover was decreased by tram-
F r e q u e n c y o f species measured on the trampled path pling impact (Fig. 4). Bryophytes and lichens were
centres c o m p a r e d with the u n t r a m p l e d surroundings at severely reduced while grass cover was m o s t reduced in
Trampling of Danish coastal vegetation 227
30
12t
.=
25 10
E
El,
•8
) 20 8F
+ g
13
i!!j
¢n
10 "6
2~
Z
0 0
Site 1 Site 2 Site 3 Site 4 Site 5 Site 1 Site 2 Site 3 Site 4 Site 5
Forbs Grasses • W o o d y sp. r~ + Trampling [] - Trampling
Fig. 2. Number of vascular plant species at five sites: site 1, Fig. 3. Diversity measured as number of species per m 2 (aver-
mown grassland; site 2, unmown grassland; site 3, artificial age based on 20 m 2) on the trampled path centre (+) com-
dune; site 4, natural saltmarsh; site 5, natural dune. The pared with the untrampled surroundings ( ) .
species are divided into forbs, graminoids (Poaceae,
Juncaceae and Cyperaceae), and woody species on the 150
trampled centre of the paths (+) compared with the untrampled
surroundings ( ) ,
120 +
& . ! ÷
the dune communities (sites 3 and 5). In contrast the J
percentage cover o f other vascular plants was not
significantly reduced (Table 3). F o r instance, Trifolium
repens and Taraxacum spp. showed higher cover values
on the trampled paths. ¢J
Figure 5 shows the percentage distribution of life- 30
forms in the five sites (untrampled surroundings). It
appears that the vegetation of the grasslands as well as
the salt m a r s h is strongly d o m i n a t e d by hemicrypto- 0
Site 1 Site 2 Site Site 4 Site 5
phytes. In the artificial dune, annuals (therophytes) are • Lichena [ ] Bryophytes
[ ] Grasses -7 Other vncular plants
the prevalent life-form, whereas in the natural dune the
therophytes, hemicryptophytes and geophytes are Fig. 4. Percentage cover of lichens, bryophytes, graminoids
almost equally important. The category 'others' (which (Poaceae, Juncaceae and Cyperaceae), and other vascular
includes c h a m a e p h y t e s and phanerophytes) was only plants on trampled path centres (+) and untrampled
represented on the sites in the m a n - m a d e K o g e Bay surroundings (-).
Seaside Park. The n u m b e r o f therophytes was strongly
80
reduced and there was also a significant reduction in
hemicryptophytes. The n u m b e r o f geophytes increased
80 e
Table 3. Results of a t-test based on the differences between
trampled path and untrampled surroundings, calculated from
the everage and standard deviation of the five sites
40 e
Increase/ t-test, d.f.=4 o
u
decrease a 5
Total number of species ( ) 2.4650 * 20
Diversity (species per m 2) (-) 2.1738 *
Number of graminoids (+) 1.6771 ns
Number of forbs ( ) 3.1534 * 0
Number of therophytes (-) 2.9418 * Site 1 Site 2 Site 3 Site 4 Site 5
Number of hermicryptophytes (-) 6.7083 ** Therophytes [] Hemieryptophytea
Number of geophytes (+) 1-1181 ns [] Geophytes • Others
Number of other life-forms (-) 0.6594 ns
Percentage cover of graminoids (-) 3.8351 ** Fig. 5. Percentage distribution of the life-forms (according to
Percentage cover of other species (-) 0.6416 ns Raunki~er, 1934) in the five communities in the untrampled
situations. The category 'others' includes chamaephytes and
"+, overall increase as a result of the tramplings; -, overall phanerophytes. The numbers above the columns indicate the
decrease. number of species in each category.
228 Ulla Vogt Andersen
'°I would cause complete loss of vegetation in both dunes
and salt marsh; 2889 passages were found to reduce the
vegetation cover of Ammophila-dominated dunes by
60-
12 50% (Boorman & Fuller, 1977).
16 10
8 Like any other disturbance, trampling can create
g
¢p
*°i open spaces for new species to become established
(Sousa, 1984). For example, Spergularia marina, an an-
nual which is unable to establish in a dense, saturated
community, was only recorded on the path in the salt
marsh. Similarly some species, such as Glaux maritima,
i,
2O cannot persist in untrampled situations where faster-
growing more competitive species tend to dominate
(Burden Randerson, 1972).
0 o 1 Some vascular plants, such as Trifolium repens, can
Site 1 Site 2 Site 3 Site 4 Site 5
withstand light trampling and take advantage of the
Vulnerable [] Favoured [] Indifferent
suppression of more sensitive neighbours by forming
Fig. 6. All species were divided into three categories, i.e. vul- dense mats as on path 2. On the other hand, both
nerable, favoured, or indifferent. Their percentages in the un-
trampled situation of the five communities were recorded. lichens and bryophytes appear to be very vulnerable to
Numbers above the columns indicate the actual number of trampling, as found by Hylgaard (1980) and Bowles
species in the category. and Maun (1982). The reason is that these plants break
easily under dry conditions and trampling in coastal
ecosystems is often more prevalent in dry weather.
on the trampled paths in all communities, but this in- Lichens are hardly ever able to regenerate after tram-
crease was not significant (Table 3). pling, as they are dependent on organic matter
All species are classified as vulnerable, favoured, or (Johnsen & Sochting, 1993), which is also reduced by
indifferent according to their scores in the )(2-test wear.
(Table 1), or their relative occurrences on either the Species like Trifolium arvense, Viola tricolor and
paths or the untrampled surroundings if their numbers Cerastium semidecandrum are easily damaged mechani-
are too low for the test to be valid (Table 2). There ap- cally, because their delicate, erect stems dry out early in
peared to be a significantly lower number of occur- the season. As they possess no buds from which they
rences on the path centre of all species in the can regenerate, they are very vulnerable to trampling.
'vulnerable' group, whereas species in the 'favoured' It is often considered that rosette plants can withstand
group showed a strong significance for occurrence on trampling impact (Bates, 1935; Burden & Randerson,
the trampled path. Figure 6 shows the distribution of 1972), but the present study indicates that some rosette
species in the three categories in untrampled surround- species are vulnerable, for instance Hypochoeris radi-
ings at the five study sites. Vulnerable species cata and Leontodon autumnalis.
contributed the majority in the dunes (sites 3 and 5)
and in the unmanaged grassland (site 2). The mown Grassland and salt marsh
grassland (site 1) had an almost equal number of After 15 years the grasslands in Koge Bay Seaside Park
vulnerable and indifferent species, while the salt marsh consist mainly of naturally colonized species (Hansen &
(site 4) was dominated by indifferent species. All five Vestergaard, 1986). They differ from the natural salt
sites were tested for differences in number of vulnerable marsh, because they are never flooded and therefore
and resistant (-- favoured + indifferent) species. The lack typical halophytic vegetation. The grasslands con-
salt marsh, site 4, differed significantly from the un- tain a lower number of resistant geophytes and a higher
mown grassland, site 2 (G 2 -- 5.718, d.f. -- 1, p -- 0-017 number of vulnerable therophytes compared with the
*), and the dune communities, site 3 (G 2 -- 6.325, d.f. = salt marsh and therefore appear to be significantly
1, p = 0.012 *), and site 5 (G ~ = 4.794, d.f. -- 1, p = more vulnerable to the impact of human trampling.
0-029 *). Although the mown grassland (site 1) and the
unmown grassland (site 2) are on similar soils, have the
DISCUSSION same initially sown species, and have been open to col-
onization from the same sources, they are quite differ-
The trampling in the five communities (estimated to be ent as a result of management. The mown grassland
1815-3630 passages per year) can be considered as very possesses fewer species overall and a lower species
light, even though it creates paths and reduces the veg- diversity, but more species indifferent to trampling than
etation cover and species diversity, because the paths the unmown grassland. The vegetation structure and
retain a persistent vegetation. The actual use of the the floristic composition of both trampled and untram-
paths, i.e. the number of visitors per year, is very diffi- pied situations in site 1 are very similar to those of the
cult to measure, as it varies throughout the year and is trampled situation at site 2. Species which are favoured
largely dependent on weather conditions. According to by trampling are to some extent also favoured by mow-
Burden and Randerson (1972) 7500 passages per year ing (Grime, 1979; Bakker, 1985). The mown grassland
Trampling o f Danish coastal vegetation 229
does not differ significantly in vulnerability to tram- dynamic stage, even if it also changes the vegetation
pling from the natural salt marsh, which has a naturally composition and decreases the species diversity in the
resistant vegetation. It appears that mowing once a more vulnerable communities. In the artificial coastal
year can change the vegetation towards a more resis- area the landscape configurations are square, linear and
tant type, but mowing has not increased either the straight, and creation of paths has provided a more
number of grasses or the total vegetation cover. Both natural appearance to the landscape. A controlled,
these features are believed to contribute to increasing light trampling impact, of for instance 5-10 visitors per
the resistance of a community (Hylgaard & Liddle, day, may contribute to securing the continuation of an
1981; Bakker, 1985). open vegetation.
Human trampling at the present levels involves no
Dunes serious detrimental effects on the five communities
The dunes appear to be the most vulnerable of the studied. If, however, the number of visitors should
communities studied in respect to vegetation cover, as increase, a loss of important species, destruction of veg-
they have a high percentage of sensitive species and few etation cover and erosion are likely, and impoverish-
that benefit from trampling. Vehicle traffic affects ment of the landscape and recreational values may
dunes more than grasslands with respect to vegetation result. Management can improve the recreational carry-
cover (Hosier & Eaton, 1980). No significant differ- ing capacity of the area, but even low-intensity man-
ences in vulnerability were found between the artificial agement can change the original vegetation and
dunes of Koge Bay Seaside Park and the natural dunes landscape morphology.
of ~Olsemagle Revle. In both sites the trampled area was Artificial coastal areas may relieve some of the pres-
characterized by geophytes typical of the embryonic or sure on the natural beaches, but creation of man-made
mobile dunes. In contrast the surrounding stabilized coastal ecosystems cannot solve the problems of wear.
dunes had a more dense vegetation cover, many annual Building of artificial coastal areas is rarely possible,
species, and a similar low percentage of open ground as and can never of course compensate for the reduction
in the mown grassland. The vegetation of embryonic of natural areas caused by human disturbance. Plant
and mobile dunes consists primarily of geophytes, able communities created on an artificial basis can be
to regenerate after natural erosion processes as well as equally vulnerable to the impact of human trampling as
after trampling, in contrast to other pioneer communi- those created through natural processes.
ties where the vegetation normally consists of annuals.
The number of therophytes and 'others' is larger in sta-
ACKNOWLEDGEMENTS
bilized dunes (Raunki~er, 1934). The man-made dunes of
Koge Bay Seaside Park, which contains a higher num- I would like to thank Gitte Calov, Jorgen Jensen, and
ber of annuals and woody species, are in a more ad- Kjeld Hansen for critical and constructive comments
vanced stage of succession compared with the natural on the manuscript. Appreciation is extended to Koge
dunes. Even if ¢01semagle Revle is considerably older City municipal authorities for providing mapping mate-
than the Seaside Park the latter appears to have a rial and to the park headquarters of Koge Bay Seaside
higher 'ecological age'. Stabilized dunes are consider- Park for information on management, visitors num-
ably more vulnerable to trampling than mobile dunes bers, etc. This research was funded by a grant from the
(Hylgaard, 1981; Johnsen & Sochting, 1993). Royal Veterinary and Agricultural University.
The dunes of NW-Europe are in some places threat-
ened by eutrophication and afforestation leading to
undesired stabilization (van der Laan, 1985). Efforts REFERENCES
such as grazing are made to reverse vegetation succes- Bakker, J. P. (1985). The impact of grazing on plant commu-
sion, to stimulate sand movement and to secure the dy- nities, plant populations and soil conditions on salt
namic dune processes (van Dijk, 1992). The impact of marshes. Vegetatio, 62, 391-8.
human trampling in the sites studied helps to keep the Bates, G. H. (1935). The vegetation of footpaths, sidewalks,
cart-tracks and gateways. J. Ecol., 23, 470-87.
dunes in a young, non-stabilized stage and can be used Bayfield, N. G. (1973). Use and deterioration of some Scot-
for the conservation of mobile dunes. tish hill paths. J. Appl. Ecol., 10, 635~14.
Mobile dunes can easily regenerate after disturbance Beeftink, W. G. (1979). The structure of salt marsh communi-
as no changes are induced in the basic soil conditions. ties in relation to environmental disturbances. In Ecological
In the grasslands and salt marsh, where soil moisture processes in coastal environments, ed. R. L. Jefferies & A. J.
Davy. Blackwell Scientific Publications, Oxford. pp. 77-93.
levels are higher, trampling may result in irreversible Boorman, L. A. & Fuller, R. M. (1977). Studies on the im-
changes in the soil, and there may be problems for re- pact of paths on the dune vegetation at Winterton, Nor-
generation of the vegetation cover on compacted soil folk, England. Biol. Conserv., 12, 203-16.
after heavy trampling (Beeftink, 1979). Bowles, J. M. & Maun, M. A. (1982). A study of the effects
of trampling on the vegetation of Lake Huron sand dunes
at Provincial Park. Biol. Conserv., 24, 273-83.
CONCLUSION Burden, R. F. & Randerson, P. F. (1972). Quantitative stud-
ies of the effects of human trampling on vegetation as an
Light trampling is beneficial to seashore communities, aid to the management of semi-natural areas. J. Appl.
because it keeps the vegetation in an immature, Ecol., 9, 439-57.
230 Ulla Vogt Andersen
Carlson, L. H. & Godfrey, P. J. (1989). Human impact man- overvdgning, forvaltning og Jorskning, ed. C. H. Ovesen &
agement in a coastal recreation and natural area. Biol. P. Vestergaard. Skov-og Naturstyrelsen, Copenhagen, pp.
Conserv., 49, 141-56. 58-63 (English summary).
Chappell, H. G., Ainsworth, J. F., Cameron, R. A. D. & Kardell, L. (1978). Vegetationsslitage Katastrof eller bara
Redfern, M. (1971). The effect of trampling on a chalk offtgenhet? Sveriges Lantbruks Universitet, Avdelningen for
grassland ecosystem. J. Appl. Ecol., 8, 869-82. Landskapsv &rd, Rap., No. 12.
Edmond, D. B. (1962). Effects of treading pasture in summer Liddle, M. J. & Greig-Smith, P, (1975a). A survey of tracks
under different soil moisture levels. N . Z . J . Agric. Res., 5, and paths in a sand dune ecosystem, I. Soils. J. Appl. Ecol.,
389-95. 12, 893-908.
Frederiksen, P. (1977). Turistslitage i et klitlandskab, Skallin- Liddle, M. J. & Greig-Smith, P. (1975b). A survey of tracks
gen 1976. Geogr. Tidsskr., 76, 68-77. and paths in a sand dune ecosystem, II. Vegetation. J.
Goldsmith, F. B., Munton, R. J. C. & Warren, A. (1970). Appl. Ecol., 12, 908-30.
The impact of recreation on the ecology and amenity of McDonell, M. J. (1981). Trampling effects on coastal dune
semi-natural areas: methods of investigation used in the vegetation in the Parker River National Wildlife Refuge,
Isles of Scilly. Biol. J. Linn. Soc., 2, 287-306. Massachusetts, USA. Biol. Conserv., 21,289-301.
Gravesen, P. & Vestergaard, P. (1969). Vegetation of a Dan- Magnusson, M. (1986). M~mniskans p~verkan p~t Sandham-
ish off-shore barrier island. Bot. Tidsskr., 65, 44~69. merens dynomr~,de i syd6stra Sk~ne. Sv. Bot. Tidskr., 80,
Grime, J. P. (1979). Plant strategies and vegetation processes. 81 93 (English summary).
J. Wiley, Chichester. Malmer, N. (1974). Scandinavian approach to vegetation sci-
Hansen, K. (1981). DanskJeltflora. Gyldendal, Copenhagen. ence. Medd. Avd. Ekol. Bot.. Univ. Lund, 2.
Hansen, K & Jensen, J. (1972). The vegetation on roadsides Meijer, L. W. (1992). Recreation in the Dutch Wadden dune
in Denmark. Qualitative and quantitative composition. areas, a curse or a blessing. In Dune management in the
Dansk Bot. Ark., 28, 1 61. Wadden Sea area. Proc. Trilateral Working Conference,
Hansen, K. & Vestergaard, P. (1986). Initial establishment of 3rd, ed. G. Hilgerloh. Nordeney, Germany 8-12 September
vegetation in a man-made coastal area in Denmark. Nord. 1991, pp. 55-61.
J. Bot., 6, 479-95. Page, R. R., da Vinha, S. G. & Agnew, A. D. Q. (1985). The
Hosier, P. E. & Eaton, T. E. (1980). The impact of vehicles reaction of some sand-dune plant species to experimentally
on dune and grassland vegetation on a south-eastern North imposed environmental change: a reductionist approach to
Carolina barrier beach. J. Appl. Ecol., 17, 173 82. stability. Vegetatio, 61, 105 14.
Hylgaard, T. (1980). Recovery of plant communities on Raunki~er, C. (1934). The IifeJorms of plants and statistical
coastal sand-dunes disturbed by human trampling. Biol. plant geography. Clarendon Press, Oxford.
Conserv., 19, 15-25. Sousa, W. P. (1984). The role of disturbance in natural com-
Hylgaard, T. & Liddle M. J. (1981). The effect of human munities. Ann. Rev. Ecol. Syst., 15, 353-91.
trampling on a sand dune ecosystem dominated by Em- van der Laanl, D. (1985). Changes in the flora of the coastal
petrum nigrum. J. Appl. Ecol., 18, 559 69. dunes of Voorne (The Netherlands) in relation to environ-
Jensen, F. (1993). Dunes - - managements and threats. In mental changes. Vegetatio, 61, 87-95.
Danske klitter - - overv~gning, forvaltning og forskning, ed. van Dijk, H. W. J. (1992). Grazing dc.mestic livestock in
C. H. Ovesen & P. Vestergaard. Skov-og Naturstyrelsen, Dutch coastal dunes: experiments, experiences and perspec-
Copenhagen, pp. 25-31 (English summary). tives. In Coastal dunes, ed. Carter, R. G. W., Curtis, T. G.
Johnsen, I. & Soehting, U. (1993). Lichens dominated heath- F. & Sheehy-Skeffington, M. J. Balkema, Rotterdam. pp.
lands - - dynamics and vulnerability. In Danske klitter - - 235 50.
,~ 1995 Elsevier Science Limited
Printed in Great Britain. All rights reserved
0006-3207/'95/$09.50+.00
ELSEVIER 0006-3207(94)00031-X
RESISTANCE OF DANISH COASTAL VEGETATION TYPES TO
H U M A N TRAMPLING
Ulla Vogt Andersen
Botanical Section, Department of Botany, Dendrology and Forest Genetics, Royal Veterinary & Agricultural University,
Rolighedsvej 21, DK-1958 Frederiksberg C, Denmark
(Received 14 January 1994; revised version received 20 May 1994; accepted 8 June 1994)
Abstract 1970; Hosier & Eaton, 1980), and loss of biodiversity
Five coastal communities under influence o f human tram- (McDonell, 1981). Because the impact results in soil
pling, i.e. a natural salt marsh, a natural dune, a man- modification, the changes in vegetation composition
made dune, and two man-made coastal grasslands, were and structure can be irreversible (Beeftink, 1979). Low
studied. The vegetation of five paths (one in each com- levels of trampling can, however, have a beneficial infl-
munity) created and sustained by, human trampling was uence on species diversity (Liddle & Greig-Smith, 1975b;
analysed with respect to floristic composition and species Boorman & Fuller, 1977), by keeping the communities
cover. The total number of vascular plant species, species in a dynamic stage (Magnusson, 1986).
diversity', and the total vegetation cover were significantly Careful planning of recreational impact is necessary
reduced. The therophytes and hemicryptophytes were sig- and a knowledge of the resistance of seashore communi-
nificantly reduced, while the geophytes were indifferent to ties is an important basis for planners. The aims of the
the impact o f human trampling. Man-made and natural p, esent study were to compare the vulnerability of five
dunes were most vulnerable, coastal grasslands were different plant communities, i.e. a natural salt marsh, a
intermediate, and the salt marsh was the most resistant natural dune, a man-made dune, a man-made coastal
to human trampling. grassland under management, and a man-made coastal
grassland without management. These communities were
Keywords: human trampling, disturbances, species re- analysed with emphasis on (1) the trampling-induced
sistance, man-made ecosystems, coastal communities. changes in vegetation composition and species diversity;
(2) the ability of individual species to withstand injury; and
(3) the possible differences in resistance between man-
INTRODUCTION
made and natural coastal areas and between managed
For holiday makers of Northern Europe seashores rank and unmanaged plant communities. Plant nomenclature
among the most attractive areas. At present, employ- follows Hansen (1981).
ment and income in many coastal areas are largely
dependent on the tourist industry, and tourism is a growth
STUDY AREAS
sector with a large potential in economic terms (Meijer,
1992; Jensen, 1993). Tourist activities, however, can The studies were made in Denmark in Koge Bay Sea-
have severe effects on coastal areas, and contribute to side Park and Olsemagle Revle (Fig. 1), both offshore
their destruction and reduction of their nature and barriers situated 20 km apart, on the coast of the Baltic
recreational values. A direct result of leisure activities is Sea. Olsemagle Revle was created by natural processes:
disturbance to the flora and fauna. Several researchers interaction of sand deposition and plant growth during
have studied the impact of human trampling in dunes the period 1900-1930. In contrast, Koge Bay Seaside
and other seashore plant communities. The natural re- Park was planned by architects and constructed for
sponses are soil compaction (Bates, 1935; Liddle & recreational purposes, in the years 1978-1980.
Greig-Smith, 1975a), reduction in soil organic matter Olsemagle Revle covers 200 ha and consists of a 5
(Boorman & Fuller, 1977; Hylgaard & Liddle, 1981), km sandy beach with a well-developed strandline vege-
decrease of vegetation cover (Burden & Randerson, tation, a row of dunes with a height of approximately 3
1972; Bowles & Maun, 1982), decrease in biomass pro- m, and a salt marsh with a Phragmites-Scirpus swamp
duction (Edmond, 1962; Liddle & Greig-Smith, 1975b), on the landward side of the barrier which encloses a
reduction in number of flowering species (Goldsmith et lagoon. The salt marsh has never been grazed or mown.
al., 1970; Hylgaard, 1980), disappearance of vulnerable The only man-made facility is a constructed dam with
species (Bates, 1935; Chappel et al., 1971), creation of an access road. A comprehensive study of the vegeta-
paths (Bayfield, 1973; Hylgaard & Liddle, 1981), tion was made by Gravesen and Vestergaard (1969).
erosion (Frederiksen, 1977; Carlson & Godfrey, 1989), Koge Bay Seaside Park covers 500 ha and consists
interference in the natural succession (Goldsmith et al., of 8 km of sandy beach with 3-m high artificial dunes
223
224 Ulla Vogt Andersen
1 Kege Ba' _Seaside Park
N
,,, .., t.~....o,~. .
)a~'~'~'~3~ ~ ~ Lagoon
¢ • f~.~ t';,..... ~\,.:--,~-~.,j....~,...<
~ :~ .. ~~ ,.,...~..~. ":
~,...
" ~ ' ~ Baltic Sea
LEGEND:
L~] Parking lot
[] Dunes
i~! Mainlandbehind original coastline
Road with vehicle access
Constructed path/bike lane
. . . . Path created by human trampling
Sites where surveys a r e undertaken
Study area
Fig. 1. Map showing the location of the study areas in Denmark (inset bottom left), and the position of the study sites 1-5 in
Kege Bay Seaside Park (top) and Olsemagle Revle (bottom right).
constructed on two long, narrow offshore barrier the artificial dunes Ammophila arenaria was planted in
islands. The new artificially advanced coastline is con- 1978. All other species occurring in the area in 1992
nected with the original natural coastline by bridges have been invading as a result of primary succession
and dykes. Behind the offshore barriers are six lagoons. (Hansen & Vestergaard, 1986). Several facilities for visitors
The following vegetation types and plant communities are are available: harbours, roads, parking areas, toilets, life
present: sandy beach, sand dunes, grasslands, former guards, ice-cream stands, etc. Being closer to Copen-
tidal meadows, lagoon shores, roadsides, shrubberies, and hagen city and having all the recreational facilities,
small planted groves. The grasslands were sown in 1979 Kege Bay Seaside Park is visited by more than 500,000
with a seed mixture of grasses consisting of Festuca rubra, people each year. In contrast, Olsemagle Revle receives
F. arundinaeea, Lolium multiflorum and L. perenne. In less than 100,000 visitors in the same period.
T r a m p l i n g o f Danish coastal vegetation 225
Table 1. Percentage frequency of 32 taxa including bare soil in +, trampled path centre, and -, untrampled surroundings at five sites
Site 1, m o w n grassland; site 2 u n m o w n grassland; site 3, artificial dune; site 4, natural salt marsh; site 5, natural dune, The species
are tested by likelihood-ratio X 2 for whether their occurrence is independent of trampling impact. Results are divided into groups ac-
cording to their response. G 2 -- Likelihood ratio value (d.f. -- 1) *** p < 0.001, **p < 0.01, * p < 0.05. Favoured, species occurring
only o n the paths; Vulnerable, species occurring only in untrampled situations; Indifferent, species occurring in b o t h situations.
Species Site 1 Site 2 Site 3 Site 4 Site 5 G2 Significance
+ + + + - +
Favoured
Bare soil 100 100 100 25 100 100 85 - 100 60 52.099 ***
Elytrigia repens -- -- -- 90 30 -- 30 -- 3-897 *
Festuca arundinacea 10 -- 80 5 . . . . 20.103 ***
Glaux maritima -- -- -- 55 15 -- -- 4.916 *
Leymus arenarius . . . . . . . . 75 25 5.788 *
Vulnerable
Bryophytes 70 90 65 70 -- 100 -- 20 60 85 23-971 ***
Lichens -- -- -- 5 80 -- -- 10 28.338 ***
Cirsium arvense 5 30 15 15 55 -- -- -- 13-101 ***
Cerastium semidecandrum . . . . 30 -- -- 25 15.890 ***
Erigeron acer 5 -- 5 5 5 55 -- -- -- 6.220 *
Holcus lanatus -- -- 95 . . . . 28.338 ***
Hypochoeris radicata . . . . . . . . 15 60 6-220 *
R u m e x acetosella 10 5 95 -- -- 20 50 25.034 ***
Tanacetum vulgare . . . . . . 15 80 10.700 **
Trifolium arvense 10 90 65 -- 20 I __ -- 30.969 ***
Viola tricolor . . . . . 5 65 12.978 ***
Indifferent
Agrostis stolonifera -- -- 90 90 -- -- 100 95 -- -- 0.021 ns
Ammophila arenaria -- -- 90 100 . . . . . 0.130 ns
Ammophila x Calamagrostis . . . . . 5 95 100 0-000 ns
Chamaenerion angustifolium . . . . 30 65 . . . . 2.911 ns
Dactylis glomerata 15 30 65 60 . . . . 0 142 ns
Festuca rubra I00 100 100 100 30 100 100 -- 20 0.088 ns
Festuca ovina -- -- -- 80 100 . . . . 0.543 ns
Hippopha# rhamnoides 30 -- 5 30 . . . . . 0.082 ns
Honckenya peploides . . . . 25 30 0.096 ns
Lathyrusjaponicus -- 60 45 -- -- 5 30 0.000 ns
Odontites verna . . . . . . 95 95 -- 0.00 ns
Phragmites australis 25 . . . . 55 60 -- 0.666 ns
Plantago maritima . . . . . 100 80 -- I 0-543 ns
Poa pratensis -- 10 10 15 -- 5 5 15 25 1.629 ns
Scirpus maritimus . . . . . 45 65 -- -- 0.821 ns
Taraxacum spp. 100 100 95 95 95 100 -- -- 10 10 0-021 ns
Trifolium repens 30 5 100 100 -- 5 5 -- 0-677 ns
METHODS c e n t r e o f all five p a t h s 0.6 m a p a r t , a n d a s i m i l a r set o f
q u a d r a t s p l a c e d i n t h e v e g e t a t i o n a l o n g s i d e , 2.5 m f r o m
Five paths, with a width of at least 1 m, and their sur- the centre of the path. Other studies have shown that
r o u n d i n g c o m m u n i t i e s w e r e s e l e c t e d as s u r v e y sites. people tend to follow paths very strictly (Bayfield,
Three are situated in Koge Bay Seaside Park: one on 1973; F r e d e r i k s e n , 1977; K a r d e l l , 1978; H y l g a a r d ,
t o p o f t h e a r t i f i c i a l d u n e s (site 3) a n d t w o i n t h e g r a s s - 1980), a n d t h e r e f o r e t h e v e g e t a t i o n o u t s i d e t h e p a t h s
l a n d s b e h i n d t h e d u n e s . T h e g r a s s l a n d a t J~egerso (site c a n b e r e g a r d e d as u n t r a m p l e d . A t o t a l o f 100 u n t r a m -
1) is c u t o n c e a y e a r a n d H i p p o p h a ~ r h a m n o i d e s is c o n - p i e d a n d 100 t r a m p l e d p l o t s w e r e t h u s a n a l y s e d .
t r o l l e d w i t h h e r b i c i d e s , w h i l e t h e g r a s s l a n d a t Lille In each plot the cover of individual species was mea-
Vejleso (site 2) is u n d e r g o i n g n a t u r a l d e v e l o p m e n t w i t h o u t sured using the Hult-Sernander-DuRietz scale of cover
any management. Two paths are situated on Olsemagle c l a s s e s ( M a i m e r , 1974). I n t h e s t a t i s t i c a l t r e a t m e n t o f
R e v l e : o n e o n t h e d u n e t o p (site 5) a n d o n e i n t h e s a l t data the cover values were transformed into the follow-
m a r s h (site 4). ( F i g . 1). T h e p a t h s i n K o g e B a y S e a s i d e i n g p e r c e n t a g e v a l u e s ( H a n s e n & J e n s e n , 1972): less
P a r k a r e u s e d b y a p p r o x i m a t e l y 10 v i s i t o r s e a c h d a y t h a n 1/16 = 2 % ; 1 / 1 6 - 1 / 8 = 9%; 1 / 8 - 1 / 4 = 18%; 1 / 4 - 1 / 2
(yearly average, according to the Park headquarters). = 36%; a n d m o r e t h a n 1/2 = 7 2 % . F r e q u e n c y o f species
T h e u s e o f t h e p a t h s a t t 0 1 s e m a g l e R e v l e is n o t k n o w n . w a s m e a s u r e d as t h e p e r c e n t a g e o c c u r r e n c e in t h e 1 - m 2
H o w e v e r , f r o m c o u n t s d u r i n g t h e s u r v e y it is e s t i m a t e d plots. Species diversity was measured as number of
t h a t t h e p a t h s a r e u s e d b y five v i s i t o r s d a i l y . species per plot based on an average from 20 plots.
I n J u l y 1993, 20 1 - m 2 q u a d r a t s w e r e p l a c e d a l o n g t h e P l a n t n o m e n c l a t u r e f o l l o w s H a n s e n (1981).
226 Ulla Vogt Andersen
Table 2. Percentage frequencies of 33 taxa grouped according to their relative occurrence on either trampled path (+) or untrampled
surroundings (-)
For explanation of sites and categories, see Table 1.
Species Site 1 Site 2 Site 3 Site 4 Site 5
+ - + + -- + -- +
Favoured
Lolium perenne -- 20 -- -- 5 -- -- --
Poa annua -- -- 5 . . . . . . .
Spergularia marina . . . . . 35 --
Tussilago fitrfara 5 . . . . . . .
Vulnerable
Aira praecox . . . . . . . 10
Anthyllis vulneraria -- 5 -- 5 . . . . .
Arenaria serpyllifolia . . . . 5 -- --
A triplex prostrata ssp. prostrata . . . . . . 5
Carduus crispus . . . . 5 -- -- --
Conyza canadensis -- -- -- 5 -- -- 5
Galium verum . . . . 5 -- -- --
Leontodon autumnalis . . . . . . . 10
Melilotus alba -- 5 -- 5 . . . . .
Plantago lanceolata -- --- 15 . . . .
Rosa canina -- -- 5 . . . . . .
Rumex crispus . . . . . 5 -- -- --
Tragopogon pratensis -- -- -- 10 . . . .
Trifolium dubium -- -- 30 -- --
Triglochin maritimum . . . . . . 5
Senecio vernalis . . . . 10 -- -- 20 85
Vicia sativa ssp. angusti~blia -- 10 -- 5 . . . . .
Vicia hirsuta -- -- 35 . . . . . .
Indifferent
Artemisia vulgaris 10 15 5 . . . . .
Carex arenaria . . . . 30 5 -- -- --
Cerastium fontanum ssp. triviale 5 5 -- 5 . . . . .
Elytrigia juneeiforme . . . . . 35 10
Juncus gerardi . . . . . . 15 15
Lotus tenuis . . . . . 10 25 -- --
Potentilla anserina . . . . 10 15
Rosa rugosa -- -- 5 10 30 -- --
TriJolium pratense 15 5 -- 10 . . . .
Triglochin palustre . . . . 5 15 -- --
Vicia craeca . . . . . 5 10
Species resistance to h u m a n trampling and their abil- all sites is shown in Tables 1 and 2. Figure 2 shows the
ity to regenerate is to a large extent dependent u p o n n u m b e r o f species in each plant community. The high-
the position o f their growth points and surviving buds est n u m b e r o f species (24), was f o u n d in the u n m o w n
(Burden & Randerson, 1972; Liddle & Greig-Smith, grassland o f site 2. The total n u m b e r o f species was
1975b; H y l g a a r d & Liddle, 1981). Raunki~er's (1934) reduced significantly by the trampling impact at all
life-form classification o f species is therefore used here. sites (Table 3), although in the m o w n grassland (site 1)
G r a m i n o i d s (including Poaceae, Juncaceae and Cyper- the same n u m b e r occurred in b o t h situations, and in
aceae) are treated as a special g r o u p because o f their the natural salt m a r s h (site 4) little difference was
flexible, linear leaves, and their ability to regenerate found. The reduction was m o s t noticeable a m o n g the
after trampling (Bates, 1935; Chappell e t a L , 1971; forbs, in contrast to the grasses o f which there are
Page e t a l . , 1985). generally m o r e species on the trampled path.
Paths and u n t r a m p l e d surroundings were c o m p a r e d The species diversity o f the communities was gener-
with respect to the frequency o f occurrence o f 30 ally low and the highest diversity (>8 species/m 2) was
species, together with bare soil, lichens, and bryo- f o u n d in the u n m o w n , u n t r a m p l e d grassland (site 2)
phytes, using a likelihood-ratio X2-test. The remaining (Fig. 3). The diversity decreased significantly as a result
n u m b e r o f species (22) occurred too infrequently for o f the trampling impact. The reduction was largest in
the X2-test to be valid. the natural dune and in the u n m o w n grassland,
whereas no reduction occurred on the path in the salt
marsh, site 4.
RESULTS
The total vegetation cover was decreased by tram-
F r e q u e n c y o f species measured on the trampled path pling impact (Fig. 4). Bryophytes and lichens were
centres c o m p a r e d with the u n t r a m p l e d surroundings at severely reduced while grass cover was m o s t reduced in
Trampling of Danish coastal vegetation 227
30
12t
.=
25 10
E
El,
•8
) 20 8F
+ g
13
i!!j
¢n
10 "6
2~
Z
0 0
Site 1 Site 2 Site 3 Site 4 Site 5 Site 1 Site 2 Site 3 Site 4 Site 5
Forbs Grasses • W o o d y sp. r~ + Trampling [] - Trampling
Fig. 2. Number of vascular plant species at five sites: site 1, Fig. 3. Diversity measured as number of species per m 2 (aver-
mown grassland; site 2, unmown grassland; site 3, artificial age based on 20 m 2) on the trampled path centre (+) com-
dune; site 4, natural saltmarsh; site 5, natural dune. The pared with the untrampled surroundings ( ) .
species are divided into forbs, graminoids (Poaceae,
Juncaceae and Cyperaceae), and woody species on the 150
trampled centre of the paths (+) compared with the untrampled
surroundings ( ) ,
120 +
& . ! ÷
the dune communities (sites 3 and 5). In contrast the J
percentage cover o f other vascular plants was not
significantly reduced (Table 3). F o r instance, Trifolium
repens and Taraxacum spp. showed higher cover values
on the trampled paths. ¢J
Figure 5 shows the percentage distribution of life- 30
forms in the five sites (untrampled surroundings). It
appears that the vegetation of the grasslands as well as
the salt m a r s h is strongly d o m i n a t e d by hemicrypto- 0
Site 1 Site 2 Site Site 4 Site 5
phytes. In the artificial dune, annuals (therophytes) are • Lichena [ ] Bryophytes
[ ] Grasses -7 Other vncular plants
the prevalent life-form, whereas in the natural dune the
therophytes, hemicryptophytes and geophytes are Fig. 4. Percentage cover of lichens, bryophytes, graminoids
almost equally important. The category 'others' (which (Poaceae, Juncaceae and Cyperaceae), and other vascular
includes c h a m a e p h y t e s and phanerophytes) was only plants on trampled path centres (+) and untrampled
represented on the sites in the m a n - m a d e K o g e Bay surroundings (-).
Seaside Park. The n u m b e r o f therophytes was strongly
80
reduced and there was also a significant reduction in
hemicryptophytes. The n u m b e r o f geophytes increased
80 e
Table 3. Results of a t-test based on the differences between
trampled path and untrampled surroundings, calculated from
the everage and standard deviation of the five sites
40 e
Increase/ t-test, d.f.=4 o
u
decrease a 5
Total number of species ( ) 2.4650 * 20
Diversity (species per m 2) (-) 2.1738 *
Number of graminoids (+) 1.6771 ns
Number of forbs ( ) 3.1534 * 0
Number of therophytes (-) 2.9418 * Site 1 Site 2 Site 3 Site 4 Site 5
Number of hermicryptophytes (-) 6.7083 ** Therophytes [] Hemieryptophytea
Number of geophytes (+) 1-1181 ns [] Geophytes • Others
Number of other life-forms (-) 0.6594 ns
Percentage cover of graminoids (-) 3.8351 ** Fig. 5. Percentage distribution of the life-forms (according to
Percentage cover of other species (-) 0.6416 ns Raunki~er, 1934) in the five communities in the untrampled
situations. The category 'others' includes chamaephytes and
"+, overall increase as a result of the tramplings; -, overall phanerophytes. The numbers above the columns indicate the
decrease. number of species in each category.
228 Ulla Vogt Andersen
'°I would cause complete loss of vegetation in both dunes
and salt marsh; 2889 passages were found to reduce the
vegetation cover of Ammophila-dominated dunes by
60-
12 50% (Boorman & Fuller, 1977).
16 10
8 Like any other disturbance, trampling can create
g
¢p
*°i open spaces for new species to become established
(Sousa, 1984). For example, Spergularia marina, an an-
nual which is unable to establish in a dense, saturated
community, was only recorded on the path in the salt
marsh. Similarly some species, such as Glaux maritima,
i,
2O cannot persist in untrampled situations where faster-
growing more competitive species tend to dominate
(Burden Randerson, 1972).
0 o 1 Some vascular plants, such as Trifolium repens, can
Site 1 Site 2 Site 3 Site 4 Site 5
withstand light trampling and take advantage of the
Vulnerable [] Favoured [] Indifferent
suppression of more sensitive neighbours by forming
Fig. 6. All species were divided into three categories, i.e. vul- dense mats as on path 2. On the other hand, both
nerable, favoured, or indifferent. Their percentages in the un-
trampled situation of the five communities were recorded. lichens and bryophytes appear to be very vulnerable to
Numbers above the columns indicate the actual number of trampling, as found by Hylgaard (1980) and Bowles
species in the category. and Maun (1982). The reason is that these plants break
easily under dry conditions and trampling in coastal
ecosystems is often more prevalent in dry weather.
on the trampled paths in all communities, but this in- Lichens are hardly ever able to regenerate after tram-
crease was not significant (Table 3). pling, as they are dependent on organic matter
All species are classified as vulnerable, favoured, or (Johnsen & Sochting, 1993), which is also reduced by
indifferent according to their scores in the )(2-test wear.
(Table 1), or their relative occurrences on either the Species like Trifolium arvense, Viola tricolor and
paths or the untrampled surroundings if their numbers Cerastium semidecandrum are easily damaged mechani-
are too low for the test to be valid (Table 2). There ap- cally, because their delicate, erect stems dry out early in
peared to be a significantly lower number of occur- the season. As they possess no buds from which they
rences on the path centre of all species in the can regenerate, they are very vulnerable to trampling.
'vulnerable' group, whereas species in the 'favoured' It is often considered that rosette plants can withstand
group showed a strong significance for occurrence on trampling impact (Bates, 1935; Burden & Randerson,
the trampled path. Figure 6 shows the distribution of 1972), but the present study indicates that some rosette
species in the three categories in untrampled surround- species are vulnerable, for instance Hypochoeris radi-
ings at the five study sites. Vulnerable species cata and Leontodon autumnalis.
contributed the majority in the dunes (sites 3 and 5)
and in the unmanaged grassland (site 2). The mown Grassland and salt marsh
grassland (site 1) had an almost equal number of After 15 years the grasslands in Koge Bay Seaside Park
vulnerable and indifferent species, while the salt marsh consist mainly of naturally colonized species (Hansen &
(site 4) was dominated by indifferent species. All five Vestergaard, 1986). They differ from the natural salt
sites were tested for differences in number of vulnerable marsh, because they are never flooded and therefore
and resistant (-- favoured + indifferent) species. The lack typical halophytic vegetation. The grasslands con-
salt marsh, site 4, differed significantly from the un- tain a lower number of resistant geophytes and a higher
mown grassland, site 2 (G 2 -- 5.718, d.f. -- 1, p -- 0-017 number of vulnerable therophytes compared with the
*), and the dune communities, site 3 (G 2 -- 6.325, d.f. = salt marsh and therefore appear to be significantly
1, p = 0.012 *), and site 5 (G ~ = 4.794, d.f. -- 1, p = more vulnerable to the impact of human trampling.
0-029 *). Although the mown grassland (site 1) and the
unmown grassland (site 2) are on similar soils, have the
DISCUSSION same initially sown species, and have been open to col-
onization from the same sources, they are quite differ-
The trampling in the five communities (estimated to be ent as a result of management. The mown grassland
1815-3630 passages per year) can be considered as very possesses fewer species overall and a lower species
light, even though it creates paths and reduces the veg- diversity, but more species indifferent to trampling than
etation cover and species diversity, because the paths the unmown grassland. The vegetation structure and
retain a persistent vegetation. The actual use of the the floristic composition of both trampled and untram-
paths, i.e. the number of visitors per year, is very diffi- pied situations in site 1 are very similar to those of the
cult to measure, as it varies throughout the year and is trampled situation at site 2. Species which are favoured
largely dependent on weather conditions. According to by trampling are to some extent also favoured by mow-
Burden and Randerson (1972) 7500 passages per year ing (Grime, 1979; Bakker, 1985). The mown grassland
Trampling o f Danish coastal vegetation 229
does not differ significantly in vulnerability to tram- dynamic stage, even if it also changes the vegetation
pling from the natural salt marsh, which has a naturally composition and decreases the species diversity in the
resistant vegetation. It appears that mowing once a more vulnerable communities. In the artificial coastal
year can change the vegetation towards a more resis- area the landscape configurations are square, linear and
tant type, but mowing has not increased either the straight, and creation of paths has provided a more
number of grasses or the total vegetation cover. Both natural appearance to the landscape. A controlled,
these features are believed to contribute to increasing light trampling impact, of for instance 5-10 visitors per
the resistance of a community (Hylgaard & Liddle, day, may contribute to securing the continuation of an
1981; Bakker, 1985). open vegetation.
Human trampling at the present levels involves no
Dunes serious detrimental effects on the five communities
The dunes appear to be the most vulnerable of the studied. If, however, the number of visitors should
communities studied in respect to vegetation cover, as increase, a loss of important species, destruction of veg-
they have a high percentage of sensitive species and few etation cover and erosion are likely, and impoverish-
that benefit from trampling. Vehicle traffic affects ment of the landscape and recreational values may
dunes more than grasslands with respect to vegetation result. Management can improve the recreational carry-
cover (Hosier & Eaton, 1980). No significant differ- ing capacity of the area, but even low-intensity man-
ences in vulnerability were found between the artificial agement can change the original vegetation and
dunes of Koge Bay Seaside Park and the natural dunes landscape morphology.
of ~Olsemagle Revle. In both sites the trampled area was Artificial coastal areas may relieve some of the pres-
characterized by geophytes typical of the embryonic or sure on the natural beaches, but creation of man-made
mobile dunes. In contrast the surrounding stabilized coastal ecosystems cannot solve the problems of wear.
dunes had a more dense vegetation cover, many annual Building of artificial coastal areas is rarely possible,
species, and a similar low percentage of open ground as and can never of course compensate for the reduction
in the mown grassland. The vegetation of embryonic of natural areas caused by human disturbance. Plant
and mobile dunes consists primarily of geophytes, able communities created on an artificial basis can be
to regenerate after natural erosion processes as well as equally vulnerable to the impact of human trampling as
after trampling, in contrast to other pioneer communi- those created through natural processes.
ties where the vegetation normally consists of annuals.
The number of therophytes and 'others' is larger in sta-
ACKNOWLEDGEMENTS
bilized dunes (Raunki~er, 1934). The man-made dunes of
Koge Bay Seaside Park, which contains a higher num- I would like to thank Gitte Calov, Jorgen Jensen, and
ber of annuals and woody species, are in a more ad- Kjeld Hansen for critical and constructive comments
vanced stage of succession compared with the natural on the manuscript. Appreciation is extended to Koge
dunes. Even if ¢01semagle Revle is considerably older City municipal authorities for providing mapping mate-
than the Seaside Park the latter appears to have a rial and to the park headquarters of Koge Bay Seaside
higher 'ecological age'. Stabilized dunes are consider- Park for information on management, visitors num-
ably more vulnerable to trampling than mobile dunes bers, etc. This research was funded by a grant from the
(Hylgaard, 1981; Johnsen & Sochting, 1993). Royal Veterinary and Agricultural University.
The dunes of NW-Europe are in some places threat-
ened by eutrophication and afforestation leading to
undesired stabilization (van der Laan, 1985). Efforts REFERENCES
such as grazing are made to reverse vegetation succes- Bakker, J. P. (1985). The impact of grazing on plant commu-
sion, to stimulate sand movement and to secure the dy- nities, plant populations and soil conditions on salt
namic dune processes (van Dijk, 1992). The impact of marshes. Vegetatio, 62, 391-8.
human trampling in the sites studied helps to keep the Bates, G. H. (1935). The vegetation of footpaths, sidewalks,
cart-tracks and gateways. J. Ecol., 23, 470-87.
dunes in a young, non-stabilized stage and can be used Bayfield, N. G. (1973). Use and deterioration of some Scot-
for the conservation of mobile dunes. tish hill paths. J. Appl. Ecol., 10, 635~14.
Mobile dunes can easily regenerate after disturbance Beeftink, W. G. (1979). The structure of salt marsh communi-
as no changes are induced in the basic soil conditions. ties in relation to environmental disturbances. In Ecological
In the grasslands and salt marsh, where soil moisture processes in coastal environments, ed. R. L. Jefferies & A. J.
Davy. Blackwell Scientific Publications, Oxford. pp. 77-93.
levels are higher, trampling may result in irreversible Boorman, L. A. & Fuller, R. M. (1977). Studies on the im-
changes in the soil, and there may be problems for re- pact of paths on the dune vegetation at Winterton, Nor-
generation of the vegetation cover on compacted soil folk, England. Biol. Conserv., 12, 203-16.
after heavy trampling (Beeftink, 1979). Bowles, J. M. & Maun, M. A. (1982). A study of the effects
of trampling on the vegetation of Lake Huron sand dunes
at Provincial Park. Biol. Conserv., 24, 273-83.
CONCLUSION Burden, R. F. & Randerson, P. F. (1972). Quantitative stud-
ies of the effects of human trampling on vegetation as an
Light trampling is beneficial to seashore communities, aid to the management of semi-natural areas. J. Appl.
because it keeps the vegetation in an immature, Ecol., 9, 439-57.
230 Ulla Vogt Andersen
Carlson, L. H. & Godfrey, P. J. (1989). Human impact man- overvdgning, forvaltning og Jorskning, ed. C. H. Ovesen &
agement in a coastal recreation and natural area. Biol. P. Vestergaard. Skov-og Naturstyrelsen, Copenhagen, pp.
Conserv., 49, 141-56. 58-63 (English summary).
Chappell, H. G., Ainsworth, J. F., Cameron, R. A. D. & Kardell, L. (1978). Vegetationsslitage Katastrof eller bara
Redfern, M. (1971). The effect of trampling on a chalk offtgenhet? Sveriges Lantbruks Universitet, Avdelningen for
grassland ecosystem. J. Appl. Ecol., 8, 869-82. Landskapsv &rd, Rap., No. 12.
Edmond, D. B. (1962). Effects of treading pasture in summer Liddle, M. J. & Greig-Smith, P, (1975a). A survey of tracks
under different soil moisture levels. N . Z . J . Agric. Res., 5, and paths in a sand dune ecosystem, I. Soils. J. Appl. Ecol.,
389-95. 12, 893-908.
Frederiksen, P. (1977). Turistslitage i et klitlandskab, Skallin- Liddle, M. J. & Greig-Smith, P. (1975b). A survey of tracks
gen 1976. Geogr. Tidsskr., 76, 68-77. and paths in a sand dune ecosystem, II. Vegetation. J.
Goldsmith, F. B., Munton, R. J. C. & Warren, A. (1970). Appl. Ecol., 12, 908-30.
The impact of recreation on the ecology and amenity of McDonell, M. J. (1981). Trampling effects on coastal dune
semi-natural areas: methods of investigation used in the vegetation in the Parker River National Wildlife Refuge,
Isles of Scilly. Biol. J. Linn. Soc., 2, 287-306. Massachusetts, USA. Biol. Conserv., 21,289-301.
Gravesen, P. & Vestergaard, P. (1969). Vegetation of a Dan- Magnusson, M. (1986). M~mniskans p~verkan p~t Sandham-
ish off-shore barrier island. Bot. Tidsskr., 65, 44~69. merens dynomr~,de i syd6stra Sk~ne. Sv. Bot. Tidskr., 80,
Grime, J. P. (1979). Plant strategies and vegetation processes. 81 93 (English summary).
J. Wiley, Chichester. Malmer, N. (1974). Scandinavian approach to vegetation sci-
Hansen, K. (1981). DanskJeltflora. Gyldendal, Copenhagen. ence. Medd. Avd. Ekol. Bot.. Univ. Lund, 2.
Hansen, K & Jensen, J. (1972). The vegetation on roadsides Meijer, L. W. (1992). Recreation in the Dutch Wadden dune
in Denmark. Qualitative and quantitative composition. areas, a curse or a blessing. In Dune management in the
Dansk Bot. Ark., 28, 1 61. Wadden Sea area. Proc. Trilateral Working Conference,
Hansen, K. & Vestergaard, P. (1986). Initial establishment of 3rd, ed. G. Hilgerloh. Nordeney, Germany 8-12 September
vegetation in a man-made coastal area in Denmark. Nord. 1991, pp. 55-61.
J. Bot., 6, 479-95. Page, R. R., da Vinha, S. G. & Agnew, A. D. Q. (1985). The
Hosier, P. E. & Eaton, T. E. (1980). The impact of vehicles reaction of some sand-dune plant species to experimentally
on dune and grassland vegetation on a south-eastern North imposed environmental change: a reductionist approach to
Carolina barrier beach. J. Appl. Ecol., 17, 173 82. stability. Vegetatio, 61, 105 14.
Hylgaard, T. (1980). Recovery of plant communities on Raunki~er, C. (1934). The IifeJorms of plants and statistical
coastal sand-dunes disturbed by human trampling. Biol. plant geography. Clarendon Press, Oxford.
Conserv., 19, 15-25. Sousa, W. P. (1984). The role of disturbance in natural com-
Hylgaard, T. & Liddle M. J. (1981). The effect of human munities. Ann. Rev. Ecol. Syst., 15, 353-91.
trampling on a sand dune ecosystem dominated by Em- van der Laanl, D. (1985). Changes in the flora of the coastal
petrum nigrum. J. Appl. Ecol., 18, 559 69. dunes of Voorne (The Netherlands) in relation to environ-
Jensen, F. (1993). Dunes - - managements and threats. In mental changes. Vegetatio, 61, 87-95.
Danske klitter - - overv~gning, forvaltning og forskning, ed. van Dijk, H. W. J. (1992). Grazing dc.mestic livestock in
C. H. Ovesen & P. Vestergaard. Skov-og Naturstyrelsen, Dutch coastal dunes: experiments, experiences and perspec-
Copenhagen, pp. 25-31 (English summary). tives. In Coastal dunes, ed. Carter, R. G. W., Curtis, T. G.
Johnsen, I. & Soehting, U. (1993). Lichens dominated heath- F. & Sheehy-Skeffington, M. J. Balkema, Rotterdam. pp.
lands - - dynamics and vulnerability. In Danske klitter - - 235 50.