Use of Ammophila arenaria for Dune Stabilization
in South Africa and Its Current Distribution—
Perceptions and Problems
U. M. HERTLING                          spread. A. arenaria occurs at most sites because of its prior
R. A. LUBKE*                           planting. The South African climate appears to affect its
Department of Botany                       vigor. However, concern about the use of the alien grass has
Rhodes University                         been raised since it has proved to be a highly invasive spe-
6140 Grahamstown, South Africa                  cies in other parts of the world, particularly along the North
                                 American west coast and in Tasmania. While the CSIR
ABSTRACT / This paper examines the use of the European
                                 (Council of Scientific and Industrial Research) promotes its
dune pioneer plant Ammophila arenaria (marram or Euro-
                                 use, CNC (Cape Nature Conservation) follows a policy that
pean beach grass) for dune stabilization in South Africa in
the past and present, its present distribution in South Africa,  restricts the use of any alien plant, including A. arenaria, and
and the perceptions of coastal management agencies and      requests further research on its invasive properties. Al-
the public about its further use. The planting of A. arenaria   though a questionnaire survey shows that stabilization sites
became the most important means of dune stabilization, by     featuring large areas of A. arenaria are accepted by the
human intervention, along the South African Cape coast in     South African public, current coastal management practices
this century. Its modern distribution from the semiarid west   need to be analyzed critically. A thorough investigation of the
coast to the subtropical shores of the Eastern Cape extends    potential invasiveness of A. arenaria on South African
through various climatic zones. Although historical data are   coastal dunes will be essential and shed new light on the
missing for some areas, there is no indication of its unaided   American A. arenaria problem.




Ammophila arenaria as Primary Agent for              inhospitable area of shifting sands difficult to cross and
Dune Stabilization in South Africa                even more difficult to develop.
                                   European settlement in South Africa led to both the
  More than 80% of the South African coastline          further spread of driftsands as well as the need for their
consists of sandy beaches backed by dunes (Tinley         stabilization. The history of stabilization practices in
1985), which are largely unvegetated and mobile. Coastal
                                 South Africa has been dealt with by several authors
driftsands usually originate in dune fields where the
                                 (Braine 1903, Keet 1936, King 1939, Walsh 1969, Shaugh-
equilibrium between open sand, the natural dune
                                 nessy 1980, Stehle 1982, Avis 1989). The use of
vegetation, and the forces of nature have been dis-
                                 European marram or beach grass [Ammophila arenaria
turbed (Stehle 1987). Such disturbance can be caused
                                 (L.) Link] for dune stabilization was initiated in
by natural agents such as sea currents, storms, and
                                 the 1870s in the vicinity of Cape Town. After many
changes in climatic patterns, or through human activi-
ties such as the removal of vegetation for firewood or       trials with the grass, which had by then proved to be a
thatching, inconsiderate development, and trampling        capable sand binder in Europe and other parts
of vegetation by stock animals, humans, or off-road        of the world, it was finally agreed that the planting
vehicles. The sandy Cape Flats to the east of Cape Town,     of A. arenaria was the best method for temporary
for example, used to be vegetated and comparatively        dune stabilization in South Africa (Heywood 1894,
stable. However, following the first European settlement      Hutchins 1901, Lubke and Hertling 1995). Since 1896,
in Cape Town in the 17th century, the natural vegeta-       A. arenaria has been planted on a large scale in the Cape
tion of the Cape Flats was rapidly destroyed by wagons,      province of South Africa (Stehle 1982). Artificial dune
trampling, and overgrazing, and they turned into an        stabilization with A. arenaria is still a common practice
                                 along the entire Cape coast, mainly to prevent sand
                                 encroachment of existing development, but also to
KEY WORDS: Ammophila arenaria; Marram; European beach grass;
      Dune stabilization; South Africa; Coastal management  make new development such as hotels and holiday
                                 resorts feasible (McGwynne and others 1993, Hertling
*Author to whom correspondence should be addressed.        1997).

Environmental Management Vol. 24, No. 4, pp. 467–482                    1999 Springer-Verlag New York Inc.
468      U. M. Hertling and R. A. Lubke




Adverse Impacts of Ammophila arenaria             Hertling 1995, Hertling 1997). In this paper, we exam-
in Other Parts of the World                  ine the use of A. arenaria in South Africa in the past and
                               present and compare sites of its use with those of its
  A. arenaria has proved to be a highly invasive species  current distribution in order to establish the possible
along the west coast of North America, where it was
                               spread of the alien grass. Perceptions of South African
introduced in the mid-19th century. It appears to
                               people about stabilization sites involving A. arenaria are
induce changes in dune topography (Barbour and
                               recorded through a questionnaire at a large stabiliza-
Johnson 1977, Wiedemann 1984); it causes the decline
                               tion site at Tableview, north of Cape Town.
or even elimination of indigenous dune plants and their
communities (Barbour and others 1976, Wiedemann
1984, 1993, Boyd 1992, Pickart and others 1990, Wiede-
mann and Pickart 1996); it spreads rapidly even on      Methods
stable inland dunes (van Hook 1983, Buell and others       Information about the use of A. arenaria in South
1995) by means of sexual as well as vegetative reproduc-   Africa in past and present was obtained from past issues
tion (Baye 1990, Bencie 1990); and it bears many
                               of the Agricultural Journal of the Cape Colony and
physiological advantages that ensure its competitive
                               other literature, archive material of CNC (Cape Nature
superiority over indigenous dune plant species (Pavlik
                               Conservation), and through personal communications
1983a,b). In spite of this, A. arenaria continues to be
                               with municipalities, farmers, landowners as well as
used along the coast of California and Oregon, both by
                               government and semigovernment organizations such as
public agencies and private individuals and organisa-
                               the Department of Water and Forestry, CNC, and CSIR
tions (Wiedemann 1987, Wiedemann and Pickart 1996).
  Cullen (1999) describes a severe case of biological    (Council of Scientific and Industrial Research), which
invasion by A. arenaria in Tasmania, where the grass was   are or have been actively involved in the planting of
also introduced in the mid-19th century. As in Califor-    A. arenaria. The current distribution of A. arenaria along
nia and Oregon, it transforms the topography of Tasma-    the Cape coast was recorded for location and size of
nian dunes, grows well even on stable dunes further      A. arenaria-covered area during 1995 and 1996 between
inland, and was observed to spread alongshore by       St. Lambert’s Bay on the west coast to the Kei River
vegetative reproduction: Cullen (in preparation) found    mouth in the Eastern Cape. Further north or east
that on the west coast of Tasmania spread of A. arenaria   A. arenaria has never been planted—with the exception
has resulted from the development of steep transgress-    of one unsuccessful trial at Walvis Bay in Namibia (see
ing foredunes more prone to wave attack, the consequen-    below)—and is unlikely to occur due to the adverse dry
tial release of tillers into the sea, and the tillers being  and hot climate. Preliminary survey results were pub-
washed ashore further along the coast where they would    lished by Lubke and Hertling (1995), but the present
form vigorous new colonies. For economic and environ-     results yield more detailed information. The occur-
mental reasons, A. arenaria has not been planted in      rence of A. arenaria along the Cape coast was put in
Tasmania since 1997 (M. Pemberton personal commu-       relation with mean annual rainfall and temperatures of
nication 1998). Adverse impacts of A. arenaria on dune    the relevant areas, using climate data from Heydorn
biota and dune geomorphology have also been re-        and Tinley (1980).
ported in mainland Australia (Mitchell 1974, Heyligers      To gain information on people’s perception of
1985, Sacheti and Scott 1986), where the grass is still    stabilization sites, and of A. arenaria as the predominant
planted in smaller amounts (Druett 1991, G. Druett      plant in these, we conducted a survey at the large
personal communication 1997), and in New Zealand       stabilization site of Tableview north of Cape Town
(Esler 1970, Johnson 1982, 1993, Smith and others       (Figures 1 and 2). Between 1991 and 1994, some 4 km
1985), where the National Parks Board has ruled out      of the Tableview coastline were stabilized with A. are-
the use of A. arenaria in its reserves (Johnson 1982), but  naria at a width of 50 m, resulting in about 20 ha of
where it may still be used in other areas.
                               stabilized dunes (Figure 2). The area is fenced off, but
                               wooden footpaths give access to the beach. In April
                               1997, one hundred people were interviewed at Table-
Objectives of the Present Study                view on their views about A. arenaria and the surround-
  Concern regarding the further use of the grass for     ing stabilization site. The questionnaire (Appendix 1)
dune stabilization in South Africa was raised in 1994,    was undertaken as a random sampling, although we
and research about its potential invasiveness in South    ensured that people of different social and racial
African coastal dune systems was initiated (Lubke and     backgrounds were interviewed.
                                     Dune Stabilization in South Africa        469




Figure 1. Present distribution of A. arenaria along the South African coastline. Every known site of occurrence of A. arenaria is
shown and its approximate cover with the grass illustrated (in hectares).


Results                              Today there is no new A. arenaria stock being intro-
                                 duced to South Africa. The country contains large
  Use of Ammophila arenaria for Dune Stabilization        enough A. arenaria populations to supply its own need
  in South Africa in the Past and Present            for stabilization projects. For many of the recent projects,
  The use of A. arenaria for dune stabilization spread     the grass was obtained, for example, from the extensive
rapidly from the Cape Flats area near Cape Town with       Koeberg stabilization site on the west coast (see
the primary sites of Bellville and Eersterivier along the     Table 1).
coast, both to the north and east. From 1900, dune
stabilization was carried out on such a large scale that by      Present Distribution of Ammophila arenaria in
1909 an area of 3130 ha, and by 1934 an area of 66,886        South Africa
ha had been stabilized in the Cape Colony (Stehle 1981,       A. arenaria occurs today along some 1300 km of the
Avis 1989). The northernmost area in South Africa         South African Cape coast from the semiarid Langebaan
where A. arenaria was ever used is Elandsbaai on the west     area on the west coast to subtropical Gonubie near East
coast. The easternmost known stabilization site involv-      London (Figure 1). It is widespread in the winter
ing A. arenaria is at Gonubie near East London.          rainfall area of the Western Cape, but grows well
A. arenaria was also tried unsuccessfully at Walvis Bay,     enough in the drier and warmer climates of the bimo-
then South West Africa, in the early 1970s (Le Roux        dal rainfall area of the Southern Cape. The Eastern
1974).                              Cape is also characterized by bimodal rainfalls, but the
  All known stabilization areas in South Africa are       overall climate becomes increasingly subtropical, and
listed in Table 1, although much data on the use of        A. arenaria occurs in much smaller numbers here. Most
A. arenaria were lost. It is not known, for example,       of the Eastern Cape sites marked in Figure 1 consist of
whether A. arenaria was repeatedly introduced to South      merely a few degenerating clumps of A. arenaria. In
Africa from European or even Australian or American        contrast, some sites in the the Southern and Western
sources, or whether the material used for the stabiliza-     Cape are covered with up to 500 ha (De Mond Nature
tion was always derived from existing local populations.     Reserve) of A. arenaria.
470      U. M. Hertling and R. A. Lubke




Figure 2. Tableview stabilization site near Cape Town in January 1996 (view to southwest towards Table Mountain). Between 1991
and 1994 about 4 km of this coastline were stabilized with A. arenaria.




  Although planted in the Eastern Cape as well as in       People’s Perceptions on the Use of
the Western and Southern Cape, A. arenaria does not         Ammophila arenaria
last as long in this area. The sites at the mouths of the      The subjects were first shown three postcards
Old Woman’s River and Mtati River in the Eastern Cape
                                 with scenic views of Table Mountain from the Table-
were stabilized in the early 1980s and today carry only a
                                 view/Bloubergstrand coastline, featuring different
few degenerating clumps of A. arenaria (Figure 3A). In
                                 plant species in the foreground (Figure 6).
contrast, the sites of Visbaai, Vleesbaai, and Garcia
                                 When asked which of these postcards would be
Nature Reserve in the Southern Cape were stabilized in
                                 their preferred choice to send to friends overseas
the 1960s and early 1970s and are still covered with
                                 as a ‘‘typical South African scenery,’’ 32% voted
vigorous A. arenaria (Figure 3B). The subtropical East-
                                 for postcard A, which featured the A. arenaria stabiliza-
ern Cape to the east of Port Elizabeth appears to be too
warm for A. arenaria, with mean annual temperatures at      tion as a foreground to Table Mountain. However,
East London and Great Fish Point of 18.7°C and 18.0°C      even more people (46%) voted for the equally
respectively (Table 2). Equally unfavorable conditions      unnatural garden community of postcard B, complete
prevail along the dry west coast, with a mean annual       with bitter aloe (Aloe ferox), which does not grow
precipitation of 229 mm at Cape Columbine (Table 2).       naturally on southwest coast dunes. Only 22% pre-
This area is characterized by a dry period from at least     ferred postcard C with a naturally occurring beach
November to April. Consequently, at the stabilization      daisy (Didelta carnosa) hummock in the foreground.
sites of Elandsbaai and Ysterfontein, both created in the    While postcards A and B were chosen mostly because
late 1970s (Table 1), no A. arenaria was observed in       of their photographic quality and layout (41% and
1995. However, even the more favorable climatic re-       24%, respectively), people voted for postcard C almost
gions of the southwestern Cape (weather stations of       certainly because of the plants in the foreground
Cape Town and Cape Agulhas in Table 2) do not receive      (82%). The indigenous Didelta carnosa was acknowl-
sufficient, continuous rainfall, and artificial watering     edged by these people as the only ‘‘real’’ west coast
remains a prerequisite for the successful long-term       plant.
stabilization of South African dunes with A. arenaria.       The stabilizing grass of postcard A and the surround-
                                      Dune Stabilization in South Africa        471


Table 1.  Stabilization of South African coastal driftsands with A. arenaria (from west to east)a
                Total area planted     Time of        Source of        Plantings carried
Site             with A. arenaria (ha)   plantings        A. arenaria         out by
Elandsbaai               —        1975–1980           —      Forestry
Saldanha—Danger Bay          0.4       1920s–1930s          —      private
Langebaan—Chemfos site         —         1996      Koeberg         Top Turf
West Coast Nt. Park—          —          —             —              —
 Schrywershoek
Ysterfontein              —        1975–1980           —      Forestry
Koeberg Nuclear Power        400         1980–1982           —      Eskom
 Station
Melkbosstrand             —         1991      Koeberg         ‘‘Water’s Edge’’ Dev., CSIR
Bloubergstrand            19         until 1918                 Cape Forestry
Big Bay                0.76       1995–1996     Koeberg         Regional Services
Tableview              20         1991–1994     Koeberg         Milnerton Municipality,
                                                   CSIR
Milnerton—Zonnekus           —          —            —      ‘‘Woodbridge Island’’ Dev.
Bellville (then Durbanroad)  sowing exp. (failure)    1876     seed from England    Cape Forestry (J. Storr Lister)
Bellville              planting      1904–1937           —      Cape Forestry
Robben Island             36        1907–1908           —      Colonial Secretary
Cape Peninsula in general        —        1907–1917           —      Cape Forestry
Hout Bay                —          1990     Koeberg         Forestry, CSIR
Kommetjie                —          1951                  Forestry
Witsandbaai              10       1996 (and earlier)  Koeberg         CNC
Glencairn                —         ca. 1988    Fishhoek         CSIR
Fishhoek                1.5      1980s, 1994–1995   Fishoek, Tableview    Local Authorities, CSIR
Muizenberg               —          1897     nurseries        Cape Forestry
Muizenberg, Seekoevlei         —        from 1936          —       Forestry
Muizenberg               —        1991–1992     Koeberg         Cape City Council, CSIR
Strandfontein, Swartklip        —          1897     nurseries        Cape Forestry
Strandfontein             95         until 1918                Cape Forestry
Macassar Beach             —        1994–1995     Koeberg         Weskaap SDR
Eersterivier         sowing exp. ( success)    1892     seed from France     Cape Forestry (A.W. Heywood)
Eersterivier           planting: 193     until 1918    nurseries        Cape Forestry
Strand                ca. 0.5      1994–1995     Koeberg         Strand Municipality
Pringle Bay              ca. 8       1970s, 1991    West Coast        Overberg District Council,
                                                   CSIR
Blesberg                —        late 1970s          —      Forestry
Middelvlei (Botriviermouth)     137         until 1917                 Cape Forestry
Walker Bay State Forest       ca. 5       1912–1920     own nurseries (2.5 ha)  Cape Forestry
Walker Bay State Forest       ca. 750      1920–1968     own material       CNC
Pearly Beach              —          1960s           —      Forestry
Uilenkraal               —        since 1950          —      Forestry
Quoin Point, Buffeljags        —        1930s–1950s          —      Forestry
Rietfontein              —          1930s           —      Forestry
Brandfontein             40          1930s           —      Forestry
Cape Agulhas             61         1901–1906           —      Public Services Dept.
Cape Agulhas—             —          1930s           —      Forestry
  Papenkuilsfontein
De Mond               ca. 900      1931–1996           —     Forestry
Waenhuiskrans—Struispoint      ca. 200        1960s           —     Forestry
Waenhuiskrans—village         —           —      De Mond        Forestry
Ryspunt, Skipskop           —        1930s–1960s        —       Forestry
De Hoop                —         from 1934        —       Forestry
De Hoop                2           1994     De Mond        CNC
Witsand (S. Cape)           5         1984–1987     De Mond        Heidelberg C.P., Forestry
Garcia State Forest          —        1970–1975         —       Forestry
Kleinjongensfontein          —        1980–1985         —       Forestry
Stilbaai               4         1901–1909         —       Cape Forestry
Stilbaai               ca. 450      1932–1951         —       Forestry
Stilbaai—Lappiesbaai         ca. 3         1994     Goukamma Nat. Reserve Local Council, CSIR,
                                                 Hydromulch
Visbaai                —         1960s          —       private?
472        U. M. Hertling and R. A. Lubke




Table 1.  (Continued)
                  Total area planted      Time of         Source of          Plantings carried
Site               with A. arenaria (ha)     plantings        A. arenaria            out by
Vleesbaai                  —          1960s            —    private?
Mosselbaai—Voorbaai             —           1990      Koeberg        CSIR
Hartenbos                 49          1904–1907         —      Railway Dept.
Hartenbos                  —           1968          —      private
Klein Brakrivier              —           1966          —      Divisional Council
Groot Brakrivier              —            —           —             —
Hersham                  ca. 1           —      De Mond?       Searles Holdings
Bothastrand                 —            —           —             —
Kleinkrantz                 —         early 1980s        —             —
Swartvlei                  —            —           —             —
Sedgefield                 ca. 6        1992–1993     Goukamma Nat. Reserve Private Dev., CSIR
Goukamma Nature Reserve           —         1920s–1970s         —             —
Plettenberg Bay—              —            —           —             —
 Keurboomsrivier
Tsitsikamma River mouth          ca. 1          1930s           —         Forestry
Oysterbay/St. Francis Bay        358          1917–1924          —         Forestry
Oysterbay/St. Francis Bay          —         since 1964          —                —
Sardinia Bay                —           —            —                —
Port Elizabeth               —         1890s–1990     Western Cape         Harbour Board, Divisional
                                                         Council, Forestry
Port Elizabeth—Cape Recife         —         1976–1977     own material         Port Elizabeth Municipality
Sunday’s River mouth            4         1960s–1977          —                 —
Alexandria Dunefield—           ca. 500       1981–1991          —          Forestry
  several areas
Diaz Cross                 —            —             —               —
Port Alfred            sowing exp. (success?)     1883            —        Port Alfred Harbour Authorities
Port Alfred                —            —             —               —
Kleinemonde                —            —             —               —
Fish River mouth—Fish           0.8           1973            —               —
  Point
Old Woman’s River mouth           —          1980            —         Forestry
Mtati River mouth              —         1978–1982          —          Forestry
Hamburg                   —           —            —              —
Gulu River mouth              —         1978–1982          —          Forestry
East London—Gonubie             —          1990s      Alexandria, De Hoop      Gonubie Municipality
aData in italics are from before 1950s. All data obtained through questionnaires sent to municipalities, CNC (Cape Nature Conservation), CSIR
(Council of Scientific and Industrial Research) and to the Department of Forestry, as well as from Keet (1936), King (1939), Walsh (1968), Stehle
(1980, 1981), McLachlan and others (1982), Lubke (1985b), Reyneke (1985), Avis (1989, 1992), van Zyl (1989), Burns and Barwell (1989), and P.G.
Reyneke (personal communication, September 1997). No data available (—).



ings was identified positively by 15% of all subjects as          it is artificially planted. The latter are also more aware of
‘‘marram.’’ However, only 6% of all people thought            the problems of invasiveness that some exotic plants
that it was probably not South African, but introduced          have created in South Africa. Furthermore, they do
from Europe. Most people (72%) did not recognize it            not use the Tableview beachfront as regularly as the
from any other site in South Africa. Although in the           people who find A. arenaria attractive, which can imply
majority, it is noteworthy that only 65% of subjects           that they are not used to its looks to the same degree.
thought that A. arenaria was actually planted at             Lastly, they tend to prefer natural beaches to built-up
Tableview, while not less than 35% were convinced             beaches.
of its occurring naturally there. Table 3 displays              Although only 58% of the interviewed people were
results of the questionnaire related to the aesthetic           able to name an exotic plant species in South Africa,
impact of A. arenaria in South Africa: People who accept         64% were aware of or had heard of the problem of
A. arenaria as an attractive plant tend to believe            invasiveness of exotic plants in this country. A. arenaria
that it occurs naturally at Tableview, while people that         was never mentioned as an exotic plant, although two
think of it as unattractive are more aware of the fact that        subjects, both longtime Tableview residents, were con-
                                  Dune Stabilization in South Africa     473




Figure 3. A: Remnants of A. arenaria at the Old Woman’s River mouth near Port Alfred, Eastern Cape (February 1996). B:
Vigorous A. arenaria at Visbaai near Mossel Bay, Southern Cape (August 1995).


cerned about the grass and commented on its growth as    disapproving of the stabilization as being undertaken
‘‘it doesn’t seem to give room for other plants’’ and    merely to give people a better view and was particularly
‘‘marram is going to be as bad as Port Jackson and will   upset about the replacement of the previously varied
spoil our beautiful coastline.’’ The latter resident was   west coast dune plant community, composed of colorful
474        U. M. Hertling and R. A. Lubke




Table 2. Mean annual temperature and rainfall data           Table 3. Selected results of a questionnaire
from weather stations along the South African Cape           at the Tableview stabilisation site (see text)
coast (data from Heydorn and Tinley 1980),                                     Of 37% who do not
and from European places in the vicinity where                          Of 63% who     find A. arenaria
A. arenaria occurs naturally (data from Kendrew                         find A. arenaria  attractive or find
1930; Geurts 1982; Harding 1998)                                  attractive    it unattractive

                   Mean annual Mean annual      Believe A. arenaria to
                   temperature precipitation      occur naturally at
Weather station             (°C)    (mm)         Tableview         42.9%        32.4%
                                    Are aware of problem
South Africa                               of invasiveness of
 Cape Columbine                             exotic plants in
  (Langebaan)a             15.1       229      South Africa        65.1%        72.9%
 Cape Town (Tableview)          17.3       627     Use Tableview beach
 Cape Agulhas (De Mond)         16.8       445      promenade
 Cape St. Blaize (Visbaai)        17.9       417      regularly         73.0%        70.3%
 Cape St. Francis (Oysterbay)      17.0       666     Use Tableview beach
 Port Elizabeth (Sunday’s                        promenade daily      42.9%        35.1%
  River Mouth)             17.3       576     Prefer natural beaches
 Great Fish Point (Port Alfred)     18.0       559      to built up beaches    55.6%        62.2%
 East London (Gonubie)          18.7       808     Prefer built up
Europe                                  beaches or like both
 Bergen                  7.1      2057      types           44.4%        37.8%
 Hamburg                 8.3       699
 Vlissingen                9.9       739
 London                  9.9       622
 Bordeaux                12.4       779     rum distichum is not robust enough for the task of dune
 Lisbon                 15.4       754     stabilization, suffering in summer even more from heat stress
 Nice                  15.1       862     than A. arenaria, whereas Ehrharta villosa grows too sparsely
 Genoa                  15.6      1321     and cannot withstand as much wind as A. arenaria (M.
aLargestA. arenaria population in vicinity (see Figure 1) is given in  Burns personal communication 1995).
parentheses.                                The use of A. arenaria is further justified by CSIR
                                    through the observation that it declines in vigor with
                                    increasing sand stability and gets replaced by indig-
plants like the hottentot fig (Carpobrotus edulis) and the
                                    enous plants within some 10 years, therefore acting as a
beach daisy (Didelta carnosa).
                                    nurse plant to indigenous species (P. Raal personal
                                    communication 1994). The same view is shared by the
Discussion                               Department of Forestry, whose policy involves the
                                    sowing of indigenous plants among new A. arenaria
  Use of Ammophila arenaria for Dune Stabilization           plantings to ensure the quick replacement of A. arenaria
  in South Africa in Past and Present                 (P. G. Reyneke personal communication 1997). Simi-
  Ammophila arenaria as ‘‘noninvasive’’ alien. A. are-        larly, in Australian dune stabilization projects, A. are-
naria is the main dune stabilizing agent in South Africa.        naria is nowadays used only in connection with indig-
It is the general belief among South Africans involved         enous dune plants and is then observed to be replaced
with dune stabilization that South African A. arenaria is        fairly well (Druett 1991, G. Druett personal communica-
a weak seeder and therefore poses an insignificant            tion 1997).
threat as an invasive species (P. Raal personal communi-
cation 1994). In fact, A. arenaria is believed to be the          The controversy
best stabilizer at present in South Africa. CSIR, a             Even though A. arenaria proves to have many positive
semi-government organization which is actively in-           characters, its use has been criticized by various individu-
volved in many recent stabilization projects (see Table         als from about the late 1980s, mainly on the grounds of
1), compared aerodynamic characteristics and associ-          its foreign origin. During the 1980s South Africa experi-
ated sand trapping effectiveness of A. arenaria with          enced a green movement, which highlighted more than
those of the indigenous dune plants sea wheat (Thinopy-         ever the negative impacts that alien species can have on
rum distichum) and pipe grass (Ehrharta villosa) and          South African ecosystems. When CNC, a government
found those of A. arenaria to be far superior (P. Raal         organization, took over many dune areas from the
personal communication 1994). Furthermore, Thinopy-           Department of Forestry in 1987, no more plantings of
                                    Dune Stabilization in South Africa      475


A. arenaria were carried out. CNC adopted as a policy      Although stabilization efforts have obviously been re-
the removal of all alien species from proclaimed nature     duced radically since the first half of this century, it is
reserves, and it was emphasized that A. arenaria is       clear from Table 1 that there is still a fair amount of it
included (G. Hellstrom personal communication 1994).
            ¨                    undertaken today, usually involving A. arenaria.
  The controversy over the use of A. arenaria for dune      Despite the efforts to control dune stabilization
stabilization in South Africa was enforced during the      practices in South Africa, stabilization projects are
international DUNES 94 conference held in South         frequently undertaken not only to prevent sand en-
Africa in January 1994. Concern about the use of        croachment on existing development, but also in order
A. arenaria has since then not only been based on the      to permit new development. Development of hotels and
mere ‘‘alienness’’ of the grass in South Africa, but also    holiday resorts close to the tideline has often led to the
on proven facts about its invasiveness in other areas of    stabilization not only of dunes but of backbeach areas as
the world (see above). In the wake of the conference,      well, causing frequent erosion of the stabilization site
CNC therefore called for further research to better       (Figure 4). A recent survey of 26 South African dun-
understand the invasive properties of marram as a        efields (McGwynne and others 1993) shows that in 18 of
matter of urgency (G. Hellstrom personal communica-
                 ¨               these (69%) stabilization efforts were in progress, mostly
tion 1994). While CNC promotes a policy that re-        with the intention of combatting natural destabilization
stricts the use of all alien plant species in South       or accommodating development. Only in one case was
Africa, including A. arenaria, the CSIR will con-        stabilization undertaken in reaction to immediate sand
tinue using A. arenaria unless research shows that       encroachment problems. However, the development of
it has adverse effects on South African coastal ecosystems.   even the smallest holiday chalet should take into ac-
  Dune stabilization under critical review. Due to the    count the natural sediment movements in the area in
unusual vigor and spread of Australian acacias such as     order to avoid sand encroachment problems a few years
rooikrans (Acacia cyclops), Port Jackson wattle (A. saligna),  later (Figure 5).
and long-leaved wattle (A. longifolia), which had been
raised since the 1930s, the South African Forestry         Present Distribution of Ammophila arenaria in
Department changed its policy in 1974 to the effect that      South Africa
‘‘the use of alien plants should be phased out and only      It appears as though A. arenaria occurs at all sites
indigenous species and the noninvasive marram grass       because of prior planting in the area. Clear historical
be used’’ (Avis 1989). In consequence, the noninvasive     data are missing in some cases: besides the fairly
A. arenaria gained importance for dune stabilization.      well-recorded stabilization efforts of government and
Since 1980 the stabilization of dunes itself has been      semigovernment organizations such as the Forestry
questioned as extremely artificial human interference      Department or the CSIR (Council of Scientific and
into natural coastal dynamics, to the extent that it was    Industrial Research), A. arenaria has also been planted
only to be carried out when farmland or settlements       by farmers and other private landowners. Records of
were immediately threatened by sand encroachment.        this use are nonexistent or unobtainable, and informa-
Driftsands were now looked upon as natural areas of       tion gained from local residents is sparse. However,
great ecological and recreational value and to be        historical and geographical research so far does not
preserved as such, rather than stabilized (Avis 1989,      indicate that unaided spread of A. arenaria into any of
Council for the Environment 1991), particularly since it    the areas shown in Figure 1 has occurred, as is so
was found that their stabilization can lead to severe      obviously the case along the west coast of North America
beach erosion as the sand supply for adjacent beaches is    (Buell and others 1995). It can be concluded with
cut off (Lubke 1985a), which is confirmed by observa-      acceptable certainty that all sites shown in Figure 1 are
tions in Tasmania (M. Pemberton personal communica-       or were stabilization sites, which is confirmed by their
tion 1998).                           close correlation with the stabilization sites listed in
  An argument for the reduction of dune stabilization     Table 1.
was also the financial aspect. Stabilization costs in the      European temperature and rainfall data from both
Alexandria dunefield near Port Elizabeth were esti-       northern and southern European stations (Table 2)
mated at R 1800/ha for the period 1981–1982; thus the      reflect an overall colder and more humid climate
stabilization of the entire dunefield would have added      in A. arenaria’s home territories. It appears as though
up to ca. R 32 million (McLachlan and others 1982).       South African areas of appropriate temperature re-
Only small-scale stabilization efforts have been under-     gimes are too dry, while those of appropriate rainfall
taken in this area since the 1960s, and most of the       regimes are too warm. In addition to unfavorable
dunefield is now a nature reserve, managed by CNC.        temperature and precipitation regimes, South African
476      U. M. Hertling and R. A. Lubke




Figure 4. Eroded stabilization sites too close to the high water mark at Sedgefield near George, Southern Cape (September
1995).



shores are subjected to higher radiation and stronger     Conclusions
winds than the European ones and, furthermore, lack
                                 Since its introduction in the second half of the 19th
regular frost periods, which have been found to en-
hance germination (Huiskes 1979, van der Putten and      century, A. arenaria has become one of the most
van Gulik 1987, van der Putten 1990).             important plant species for the artificial stabilization of
                               driftsands in South Africa. It was and is planted in South
  People’s Perceptions on the Use of             Africa to such an extent that it presently occurs from the
  Ammophila arenaria                     semiarid west coast to the lush, subtropical shores of the
  Concern about the grass or about invasiveness in     eastern region of the Eastern Cape. However, an analy-
general was unusual. A connection between A. arenaria     sis of the use of A. arenaria in the present and past in
as an exotic species and the detrimental effects that     relation to its modern distribution does not indicate any
certain other exotic plant species have shown in South    tendency of A. arenaria to spread unaided into areas in
Africa is not necessarily drawn. This unsuspicious view-   which it is not needed as a stabilizer. A reason can be
point is reflected in an investigation on perceptions of    found in the South African climate. Although A. are-
professional ecologists compared to business leaders     naria has so far been looked upon as extremely tolerant
about priorities of environmental issues in South Africa   towards a variety of different climatic conditions, the
(Preston and others 1989): 55% of the professional      South African climate is possibly one important factor
ecologists perceive the problem of invasion of South     preventing the aggressively vigorous growth of the grass
African ecosystems by exotic plant and animal species as   in South Africa that can be observed so clearly along the
‘‘critical,’’ ‘‘very serious,’’ or ‘‘serious,’’ compared to  temperate shores of north California, Oregon, and
only 13% of the business leaders. A severe lack of      Tasmania, which have cool maritime climates.
information on ecological issues such as invasion be-       Although the current practice of stabilizing drift-
comes apparent among nonprofessionals. Environmen-      sands with A. arenaria appears to be widely accepted by
tal awareness in South Africa is closely connected to the   South Africans, it is advisable to exercise more caution
level of education. However, some issues—like that of     regarding the use of A. arenaria in South Africa. The
biological invasion—appear to be hardly known at all     grass is presently the primary and in most cases the only
outside a small circle of professional ecologists.      plant in South African stabilization projects. Consider-
                                    Dune Stabilization in South Africa      477




Figure 5. Driftsands threatening an inappropriate resort development at Oysterbay near Cape St. Francis, Eastern Cape (July
1995).



ing that 69% of South African dunefields examined by      as will aspects of the community and population biology
McGwynne and others (1993) were affected by develop-      of the grass (Hertling 1997). The establishment of a
ment of houses, roads, car parks, camping facilities,     potential for invasion of A. arenaria on South African
sewage pipelines and other human-induced structures,      Cape coastal dune systems will not only evaluate current
it becomes obvious that South African dunefields are      coastal management practices in South Africa but also
severely threatened by human expansion. In the long      further our knowledge about the biology of this world-
run, the stabilization of many dunes will be unavoid-     wide-used European dune pioneer species and on its
able. Therefore, a critical analysis of present stabiliza-   possible invader qualities. It will therefore shed new
tion methods and their improvement should be tackled      light on the American A. arenaria problem.
sooner rather than later.
  Despite the amount of information on the adverse
impacts of A. arenaria along the North American west      Acknowledgments
coast, a controversy about the further planting of
A. arenaria between different environmental manage-        Mike Burns and Peter Raal of the Council of Scien-
ment agents similar to the controversy in South Africa     tific and Industrial Research (CSIR) in Stellenbosch
has not taken place in California and Oregon. A.        and Peter Reyneke of the Department of Water and
arenaria continues to be used, if on a smaller scale, and   Forestry in Nelspruit provided information on the past
it is likely to continue to spread since eradication efforts  and current use of marram grass for dune stabilization
are extremely time- and labour/cost-intensive and possi-    in South Africa. Garry Druett of the Department of
bly restricted to nature reserves (van Hook 1983,       Land and Water Conservation, Nowra, New South
Wiedemann 1987, Wiedemann and Pickart 1996, Pick-       Wales, Australia, provided information on the use of
art and Sawyer 1999). Although a lot of research has      marram in Australia. Michael Pemberton of the Depart-
been done to find reasons for A. arenaria’s invasiveness    ment of Environment and Land Management in Ho-
in California and Oregon, many questions remain        bart, Tasmania, is thanked for reviewing this manuscript
open. Research topics such as the reproductive behav-     and for much information on marram grass in Tasma-
ior of A. arenaria, its genetics, and interactions with soil  nia. Susan Ziehl and Simon Walsh assisted with the
pathogens will be addressed in the South African study,    questionnaire. Mark van Aardt and Des Kleineibst are
478       U. M. Hertling and R. A. Lubke




thanked for permission to reproduce their postcards of       Hertling, U. M. 1997. Ammophila arenaria (L.) Link (marram
Table Mountain views.                        grass) in South Africa and its potential invasiveness. PhD
                                   thesis. Rhodes University, Grahamstown, South Africa,
                                   279 pp.
Literature Cited                          Heydorn, A. E. F., and K. L. Tinley. 1980. Estuaries of the Cape;
                                   part I: synopsis of the Cape coast—natural features, dynam-
Avis, A. M. 1989. A review of coastal dune stabilisation in the   ics and utilisation. National Research Institute for Oceanol-
 Cape Province of South Africa. Landscape and Urban Plan-      ogy, CSIR, Stellenbosch, 97 pp.
 ning 18:55–68.                          Heyligers, P. C. 1985. The impact of introduced plants on
Avis, A. M. 1992. Coastal dune ecology and management in the     foredune formation in south-eastern Australia. Proceedings,
 Eastern Cape. PhD thesis. Rhodes University, Grahamstown,     Ecological Society of Australia 14:23–41.
 South Africa, 300 pp.
                                  Heywood, A. W. 1894. Sand-stay grasses: marram grass—
Barbour, M. G., and A. F. Johnson. 1977. Beach and dune.       Ammophila arundinacea. Agricultural Journal of the Cape Colony
 Pages 223–261 in M. G. Barbour and J. Major (eds.),        15(7):342–343.
 Terrestrial vegetation of California. John Wiley & Sons, New
                                  Huiskes, A. H. L. 1979. Biological flora of the British Isles—
 York.
                                   Ammophila arenaria (L.) Link. Journal of Ecology 67:363–382.
Barbour, M. G., T. M. De Jong, and A. F. Johnson. 1976.
 Synecology of beach vegetation along the Pacific coast of the   Hutchins, D. E. 1901. Successful sand drift fixing at Robertson.
 United States of America: a first approximation. Journal of     Agricultural Journal of the Cape Colony 11(19):729–730.
 Biogeography 3:55–69.                       Johnson, P. N. 1982. Naturalised plants in southwest South
Baye, P. R. 1990. Comparative growth responses and popula-      Island, New Zealand. New Zealand Journal of Botany 20:131–
 tion ecology of European and American beachgrass (Ammo-       142.
 phila spp.) in relation to sand accumulation and salinity.    Johnson, P. N. 1993. Dry coastal ecosystems of New Zealand.
 PhD thesis. University of Western Ontario, London, On-       Pages 197–221 in E. van der Maarel (ed.), Ecosystems of the
 tario.                               world, 2B. Dry coastal ecosystems—Africa, America, Asia
Bencie, R. 1990. Investigation of seedling emergence and seed     and Oceania. Elsevier, Amsterdam.
 bank of Ammophila arenaria. Unpublished report. The Na-      Keet, J. D. M. 1936. Report on driftsands in South Africa.
 ture Conservancy, Arcata, California, 19 pp.            Bulletin No. 172. Department of Agriculture and Forestry,
Boyd, R. S. 1992. Influence of Ammophila arenaria on foredune     Pretoria, 44 pp.
 plant microdistribution at Point Reyes National Seashore,     Kendrew, M. A. 1930. The climates of the continents, 2nd ed.
 California. Madrono 39(1):67–76.
          ˜                         Oxford University Press, Oxford.
Braine, C. D. H. 1903. Reclamation of driftsands in the Cape    King, N. L. 1939. Reclamation of the Port Elizabeth driftsands.
 Colony. Agricultural Journal of the Cape Colony 2(23):161–178.   South African Forestry Journal 2:1–10.
Buell, A. C., A. J. Pickart, and J. D. Stuart. 1995. Introduction  Le Roux, P. J. 1974. Establishing vegetation in saline soil to
 history and invasion patterns of Ammophila arenaria on the
                                   stabilise aeolian sand at Walvis Bay, South West Africa.
 north coast of California. Conservation Biology 9(6):1587–
                                   Biologia Contemporanea 1(1):25–31.
 1593.
                                  Lubke, R. A. 1985a. Erosion of the beach at St. Francis Bay,
Burns, M. E. R., and L. Barwell. 1989. The reactivation of a
                                   Eastern Cape, South Africa. Biological Conservation 32:99–
 sediment pathway within an artificially stabilised headland
                                   127.
 bypass dune system. Pages 52–57 in 9th Australian Confer-
 ence on Coastal and Ocean Engineering, Adelaide. Institute    Lubke, R. A. 1985b. Survey of the State Farm 667—Huisklip.
 of Engineers, Australia, Canberra.                 Unpublished report. Department of Botany, Rhodes Univer-
Council for the Environment. 1991. A policy for coastal zone     sity, Grahamstown, South Africa, 23 pp.
 management in the Republic of South Africa; part 2:        Lubke, R. A., and U. M. Hertling. 1995. Is Ammophila arenaria
 guidelines for coastal land-users. Council for the Environ-     (marram grass) a threat to South African dunefields? Journal
 ment, Pretoria, 95 pp.                       of Coastal Conservation 1:103–108.
Cullen, P. 1999. Ammophila arenaria and Euphorbia paralis:     McGwynne, L. E., K. van Teylingen, and A. McLachlan. 1993.
 serious threats to the integrity of the South West Tasmanian    Coastal dune conservation—perceptions and needs of man-
 coastline. Occasional Paper, Tasmania Parks and Wildlife      agement authorities. Institute of Coastal Research Report
 Service. (in preparation).                     No. 32. University of Port Elizabeth, South Africa, 22 pp.
Druett, G. 1991. Earthmover’s training course, unit 20: special   McLachlan, A., P. R. Sieben, and C. Ascaray. 1982. Survey of a
 earthworks—beach dune reclamation. Soil Conservation        major coastal dunefield in the Eastern Cape. Zoology
 Service of New South Wales, Chatswood, NSW, Australia.       Department Report No. 10. University of Port Elizabeth,
Esler, A. E. 1970. Manawatu sand dune vegetation. Proceedings,    South Africa, 48 pp.
 New Zealand Ecological Society 17:41–46.             Mitchell, A. 1974. Plants and techniques for sand dune
Geurts, H. 1982. Almanak van weer en klimaat. B. V. W. J.      reclamation in Australia. International Journal of Biometeorol-
 Thieme & Cie, Zutphen.                       ogy 18(2):168–173.
Harding, M. 1998. Weer- en Klimaatsgids, 2nd ed. Uitgeverij     Pavlik, B. M. 1983a. Nutrient and productivity relations of the
Elmar B. V., Rijswijk.                        dune grasses Ammophila arenaria and Elymus mollis. I. Blade
                                      Dune Stabilization in South Africa         479


 photosynthesis and nitrogen use efficiency in the laboratory   Stehle, T. C. 1982. Die ontwikkeling van waaisandherwinning
 and field. Oecologia 57:227–232.                  in Suid-Afrika. Pages 181–185 in J. V. Jordaan (ed.), Saasveld
Pavlik, B. M. 1983b. Nutrient and productivity relations of the   No. 50. Directorate of Forestry, Pretoria.
 dune grasses Ammophila arenaria and Elymus mollis. II. Growth  Stehle, T. C. 1987. Sand dune ecology and driftsand reclama-
 and patterns of dry matter and nitrogen allocation as       tion. Pages 563–570 in K. von Gadow, D. W. van der Zel, A.
 influenced by nitrogen supply. Oecologia 57:233–238.        van Laar, A. P. G. Schonau, H. W. Kassier, P. W. Warkotsch,
                                              ¨
Pickart, A. J., and Sawyer. 1999. Ecology and restoration of     H. F. Vermaas, D. L. Owen, and J. V. Jordaan (eds.), Forestry
 northern California coastal dunes. California Native Plant     handbook. Southern African Institute of Forestry, Pretoria.
 Society, Sacramento, California (in press).           van der Putten, W. H. 1990. Establishment of Ammophila
Pickart, A. J., D. R. Brown, and W. Avery. 1990. Experimental    arenaria (marram grass) from culms, seeds and rhizomes.
 eradication of European beachgrass (Ammophila arenaria),      Journal of Applied Ecology 27:188–199.
 Humboldt County, California. Unpublished report. The
                                  van der Putten, W. H., and W. J. M. van Gulik. 1987.
 Nature Conservancy, Arcata, California, 12 pp.
                                   Stimulation of vegetation growth on raised coastal fore-
Preston, G. R., R. F. Fuggle, and W. R. Siegfried. 1989.       dune ridges. Netherlands Journal of Agricultural Science 35:198–
 Environmental issues in South Africa: perceptions of profes-    201.
 sional ecologists and business leaders. South African Journal
                                  van Hook, S. S. 1983. A study of Ammophila arenaria (L.) Link:
 of Wildlife Research 19(4):145–155.
                                   control methods and a management plan for the Lanphere-
Reyneke, P. G. 1985. Die herwinning van waaisand op Walker      Christensen Dunes Preserve, The Nature Conservancy.
 Bay—Staatsbos, Hermanus, langs die suidwes-kaapse kus       Unpublished report. The Nature Conservancy, Arcata, Cali-
 deur van Ammophila arenaria (L.) Link gebruik te maak. MSc     fornia, 51 pp.
 thesis. Stellenbosch University, South Africa.
                                  van Zyl, J. P. 1989. Herwinningswerk op waaisandduine in
Tinley, K. L. 1985. Coastal dunes of South Africa. South
                                   Humansdorp-distrik. Dohne Agric 11(2):8–11.
 African National Scientific Programmes Report No. 109.
 Foundation for Research Development, Pretoria, 300 pp.      Walsh, B. N. 1968. Some notes on the incidence and control of
                                  driftsands along the Caledon, Bredasdorp and Riversdale
Sacheti, U., and G. A. M. Scott. 1986. The vegetation ecology
                                  coastline of South Africa. Bulletin No. 44. Department of
 of a costal sand dune in south-eastern Australia: Gun-
                                  Forestry, Pretoria, 79 pp.
 namatta Beach. Proceedings, Royal Society of Victoria 98(2):
 73–86.                              Wiedemann, A. M. 1984. The ecology of Pacific northwest
Shaughnessy, G. L. 1980. Historical ecology of alien woody     coastal sand dunes: a community profile. US Fish Wildlife
 plants in the vicinity of Cape Town. PhD thesis. University of  Service Report FWS/OBS-84/04. Washington, DC, 130 pp.
 Cape Town (School of Environmental Studies Report No.      Wiedemann, A. M. 1987. The ecology of European beachgrass
 23), South Africa, 421 pp.                    [Ammophila arenaria (L.) Link]—a review of the literature.
Smith, S. M., R. B. Allen, and B. K. Daly. 1985. Soil-vegetation  Technical report 87-1-01. Nongame Wildlife Program, Or-
 relationships on a sequence of sand dunes, Tautuku Beach,     egon Department of Fish and Wildlife, Corvallis, Oregon,
 South-east Otago, New Zealand. Journal, Royal Society of New   18 pp.
 Zealand 15(3):295–312.                      Wiedemann, A. M. 1993. Dry coastal ecosystems of northwest-
Stehle, T. C. 1980. Waaisandbestuursplan vir die Alexandri-     ern North America. Pages 341–358 in E. van der Maarel
 akusstaatsbos. Unpublished report. District Forester, Depart-   (ed.), Ecosystems of the world, 2B. Dry coastal ecosystems—
 ment of Forestry, South Africa, 37 pp.              Africa, America, Asia and Oceania. Elsevier, Amsterdam.
Stehle, T. C. 1981. Die ontwikkelingsgeskiedenis van waaisanh-   Wiedemann, A. M., and A. Pickart. 1996. The Ammophila
 erwinning in Suid-Afrika tot en met 1909. BSc (Honours)      problem on the northwest coast of North America. Land-
 thesis. Stellenbosch University, South Africa, 30 pp.       scape and Urban Planning 34:287–299.
480       U. M. Hertling and R. A. Lubke




Appendix 1.   Questionnaire used in the survey at the Tableview stabilisation site, based on postcards
A–C (Fig. 6)
Questionnaire—beach promenade Tableview (Cape Town), April 1997
1a. Which of these three postcards do you think is most typical of S. Africa?, which one would you send to friends overseas to
  show them S. Africa?
      A          B           C
1b. Why?
  A: because of layout/composition and photographic quality of postcard
  B: because of plants in foreground of postcard
  C: both A and B
  D: other reason:
1c. Can you name any of the plants on the postcards?
                             Y          N
      yes:
1d. Do you think any of the plants on the postcards have been planted or are they natural here?
  A: planted:           B: natural:
2a. Do you think this grass (pointing at marram around) has been planted or is it natural here?
  A: planted           B: natural
2b. If A: do you know why it has been planted?
  A: beauty           B: stabilization of beach
  C: other reason:        D: don’t know
2c. Have you seen this grass anywhere else along the South African coast?
                             Y          N
2d. Do you like the look of this grass?
                     Y          N           can’t say
2e. Do you know where it comes from?
                            Y         N
2f. Have you ever read any of these signs which explain about the stabilisation?
                             Y          N
3a. Can you name an exotic plant species in S. Africa?
3b. Are you aware of the problem of invasiveness of exotic plants in S. Africa?
                             Y          N
4a. Do you live in South Africa?
                            Y         N
                      no, in:
4b. If yes, are you from
  A: the coast           B: inland
4c. If inland: do you go to the coast regularly?
                             Y          N
       how often:
4d. Do you prefer natural beaches or built up beaches like this?
  A: natural beaches          B: built up beaches
  C: depends on purpose of visit
4e. Do you use the Milnerton/Tableview beach promenade regularly?
                             Y          N
            how often:
                                     Dune Stabilization in South Africa       481




Figure 6. Three postcards on sale in Cape Town souvenir shops, showing the view from the Tableview-Bloubergstrand coastline
north of Cape Town across Table Bay with A. arenaria (A), various indigenous species in a garden (B), and the indigenous Didelta
carnosa (C) in the foreground.




Figure 6 (continued).
482      U. M. Hertling and R. A. Lubke




Figure 6 (continued).