Indian Ocean coastal belt

Full text

568

Indian Ocean Coastal Belt
W.S. Matthews
Figure 11.1 Black Rock north of Sodwana Bay on the Maputaland coast (KwaZulu-Natal)
with coastal thicket, subtropical dune forest and fragments of coastal grassland.
Ladislav Mucina, C. Robert Scott-Shaw, Michael C.
Rutherford, Kelson G.T. Camp, Wayne S. Matthews,
Leslie W. Powrie and David B. Hoare
Table of Contents
1 Introduction 570
1.1 Position and Landscape Complexity 570
1.2 Biome Identity 570
2 Ecology: Climate, Geology, Soils and Hydrology 572
2.1 Climate 572
2.2 Geology and Soils 572
2.3 Hydrology 573
3 Biogeographical Patterns: Past and Present 573
3.1 Palaeo-ecological Patterns 573
3.2 Current Biogeographical Patterns 574
4 Present Status 575
5 Threats to Natural Vegetation 575
6 Action: Conservation and Management of Resources 575
7 Further Research Challenges 575
8 Descriptions of Vegetation Units 576
9 Credits 581
10 References 581
List of Vegetation Units
CB 1 Maputaland Coastal Belt 576
CB 2 Maputaland Wooded Grassland 577
CB 3 KwaZulu-Natal Coastal Belt 578
CB 4 Pondoland-Ugu Sandstone Coastal Sourveld 579
CB 5 Transkei Coastal Belt 580
STRELITZIA 19 (2006)
569
11

570
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
by edaphically or hydrologically controlled areas of grassland,
some others have assumed that at least a significant part of
the belt had been an open to dense savanna vegetation, inter-
spersed with many areas of forest and grassland. For example,
Huntley (1984) mapped most of the IOCB as his Moist Savanna
Biome type in which was embedded smaller but significant
areas of his Lowland Forest unit. Rutherford & Westfall (1986)
mapped the IOCB as part of their Savanna Biome explicitly on
nonfloristic (only structural and climatic) criteria and followed
a defined criterion at biome scale level which precluded the
mapping of the small or narrow ‘lowland forests’. They showed
the rainfall seasonality of the IOCB as essentially summer rain-
fall tending towards even year rainfall near the northern coast.
Despite the overriding floristic affinities of forest floras within
the IOCB with the African East Coast to the north, the IOCB
(the Pondoland-Ugu Sandstone Coastal Sourveld aside) also
has some floristic links with savannas toward the northwest
interior. Examples of elements straddling both Savanna and
IOCB biomes include Acacia ataxacantha, A. caffra, A. nilotica,
A. robusta, Combretum molle, Erythrina lysistemon, Gardenia
volkensii, Gymnosporia maranguensis, Heteropyxis natalen-
sis, Ozoroa paniculosa, Sclerocarya birrea, Turraea obtusifolia
and Vangueria infausta. Acocks (1953) supported the notion
that the whole IOCB (except the Pondoland Coastal Plateau
Sourveld) ‘was naturally some form of forest’, but acknowl-
edged that the ‘veld today is a more or less open thornveld with
numerous and extensive patches of forest’.
We argue that the reasons for considering the IOCB a biome
in its own right lie in the combination of vegetation structure
and combination of climatic characters which are different
from the Savanna Biome. The ‘tropical appearance of vegeta-
tion’ (to cite the words of Bews 1920) is a result of a mixture of
growth forms such as trees, lianas and epiphytes that dominate
the zonal (forest) vegetation of the region, while grasses play
only a subordinate role in the zonal vegetation itself. Grasses
are, naturally, the structure-shaping growth form in azonal (or
intrazonal) vegetation types such as hygrophilic grasslands and
shallow-soil sourvelds, also found within IOCB. The presence of
these azonal grasslands as well as extensive secondary grass-
lands led Low & Rebelo (1996, 1998) to classify a large portion
of the IOCB as part of the Grassland Biome. The absence of
an entirely rain-free dry period (so typically characterising the
winter months in the Savanna Biome) and implied increased
probability of winter rainfall is another feature distinguishing
the IOCB from the Savanna. Although the rainfall of the IOCB
is clearly higher in summer than in winter (Figure 1), along the
Maputaland coast the winter half-year rainfall makes up as
much as 40% of the total yearly precipitation. The overall high
air-moisture saturation throughout the year gives the climate of
the IOCB its strikingly tropical character, especially in summer.
Burgess et al. (2004) classify this region as part of the global
‘Tropical Broadleaved Moist Forest’ biome—a concept still much
too broad as it would encompass not only the most (sub)tropical
forests of the Indian Ocean seaboards (spanning East London
and southern Somalia), but also afrotropical rainforests of
Central and West Africa.
The biogeographical peculiarity of the IOCB (links to inland
tropical regions of Central Africa) and regional concentration of
endemic species, initiated an evolution of ideas spanning nearly
100 years (from Marloth 1907 to Van Wyk & Smith 2001).
Marloth (1907) suggested ‘South Eastern Coast Belt’, which
would incorporate not only the immediate coastal rims of the
present KwaZulu-Natal and Transkei coast of the Eastern Cape,
but also the deep hinterland. The ‘Coast Belt’ of Bews (1920)
is a more conservative concept. In fact its extent is very close to
1. Introduction
1.1 Position and Landscape Complexity
The region covers the seaboard in two provinces of South
Africa—KwaZulu-Natal and the northern half of the Eastern
Cape (in the latter part also known as the Wild Coast). This
coastal belt in its subtropical facies extends also beyond the
national borders into Mozambique as far as the Limpopo River
mouth. It continues in strongly tropical facies further north-
wards into Tanzania, Kenya and southern Somalia (for approxi-
mate delimitation, consult Moll & White 1978: Figures 1 and 2).
In South Africa it has been known as Coastal Belt (Bews 1920)
or as Indian Ocean Coastal Belt (Moll & White 1978). It is the
region, both for a biologist and a layman, where South Africa
feels the breath of the Tropics the most.
The Indian Ocean Coastal Belt (IOCB) occurs as an almost
800 km long coastal strip between the South African border
with Mozambique as far south as the mouth of the Great Kei
River (near East London). It spans altitudes from 0–450 m (and
higher up to 600 m in the Pondoland-Ugu Sandstone Coastal
Sourveld). The landscapes of the IOCB are flat (Maputaland)
or characterised by alternating rolling hills and deeply incised
valleys (coastal stretch between Richards Bay and Port Edward
in KwaZulu-Natal and then more markedly further south to Port
St Johns as far as the Great Kei River mouth). Elevated plateaus
and deep gorges are characteristic of the Pondoland coast and
other regions with underlying sandstone geology. The belt is
about 35 km wide at some places in the north (somewhat wider
in the valley of the Thukela River), narrowing irregularly south-
wards to <20 km in parts of Pondoland to <10 km in several
parts of the Wild Coast.
The region is very densely populated, with towns such as
KwaNgwanase (Manguzi), St Lucia, Mtubatuba, KwaMbonami,
Empangeni, Richards Bay, Stanger, Tongaat, Durban
Metropolitan Area (Ethekwini), Umkomaas, Port Shepstone,
Margate and Port Edward (all KwaZulu-Natal). Further south, in
the Eastern Cape, densely populated rural areas are also found,
for example in the Xhora District. Much of the KwaZulu-Natal
coast (roughly between Stanger and Port Edward) has been
turned into a ‘South African Riviera’—an almost continuous
chain of holiday settlements frequented by millions of tourists
every year.
1.2 Biome Identity
Bews (1920: 383), in his classical paper on The plant ecology of
the coast belt of Natal established that
‘... both from the purely ecological and from the floristic stand-
points... there are good reasons why the coast belt of Natal
should be considered as part of a distinct subtropical region...’.
Later, some researchers would honour this proposition and
would recognise the peculiarity of the KwaZulu-Natal (and to
an extent also the Wild Coast) immediate seaboards as deserv-
ing special status in both ecological (Moll & White 1978; to
an extent also Huntley 1984) and biogeographical (Van Wyk
& Smith 2001, Heijnis 2004, Kirkwood 2004) terms. The over-
whelmingly large extent of transformation of the coastal belt
outside the existing strips and patches of embedded forest rep-
resents significant loss of evidence of its prior condition. This
has led to a diversity of opinions on the broad-level classifica-
tion of the vegetation of this coastal belt or its components (see
Rutherford & Westfall 1986: Table 1 and Figure 4). Although
the present work broadly follows Bews (1920) and Moll &
White (1978) who assume a dominant forest cover interrupted

STRELITZIA 19 (2006)
571
Indian Ocean Coastal Belt
the mapped extent of the IOCB in this chapter. Acocks’s (1953;
see Acocks Map No. 4 in Acocks 1988) ‘Forest and Scrubforest’
basically copies the extent of the IOCB as it is depicted in our
map. Acocks (1988) recognises the identity of the Coastal Belt
in his Veld Type 1 called ‘Coastal Forest and Thornveld’—a unit
reaching beyond East London westwards as far as Keiskamma
and incorporating most of the Maputaland coastal plains. The
combination of Low & Rebelo’s (1996, 1998) Vegetation Type
23 (Coastal Bushveld/Grassland, largely corresponding to the
Acocks’s Veld Type 1) and Vegetation Type 48 Coastal Grassland
(incorporating the Acocks’s Veld Type 3: Pondoland Coastal
Plateau Sourveld) roughly corresponds to the IOCB as featured
in this chapter. To date, Camp (1999a, b) has undertaken the
most detailed climatic, geological and pedological delimitation
of the Coastal Belt and classified it as his BRG 1 Moist Coast
Forest, Thorn and Palm Veld. We follow (within the borders of
the KwaZulu-Natal Province) this delimitation.
The World Wildlife Fund recognises five ‘ecoregions’ (Goldberg
& Frank 2004, Heijnis 2004, Kirkwood 2004, Schipper & Burgess
2004a, b) within so-called ‘Tropical and Subtropical Moist
Broadleaf Forests’ which straddle the Indian Ocean seaboards
from the Jubba River in southern Somalia as far south as Port
to intensive human influence dating since the Iron Age, but
it demonstrates (under the climatic conditions of the current
Holocene Interglacial) its clear tropical affinity.
The IOCB (or Subtropical Coastal Forest Biome) encompasses
subtropical forest (Table 11.1) as zonal vegetation, accompa-
nied by a series of intrazonal (edaphic grasslands) and azonal
vegetation types. Within the latter, we count the azonal
forest types (Northern Coastal Forest, Mangrove Forest, Swamp
Forest, Lowveld Riverine Forest) and azonal nonforest vegeta-
tion units such as Subtropical Freshwater Wetland, Subtropical
Dune Thicket and Subtropical Coastal Vegetation. For practi-
cal purposes we have singled out the forest vegetation (both
zonal and azonal) of the IOCB and feature it, alongside other
forest types, within Chapter 12. The nonforest vegetation is
the subject of Chapters 13 and 14. The vegetation units in
this particular chapter include the edaphic grassland units
(Maputaland Wooded Grasslands, Pondoland-Ugu Sandstone
Coastal Sourveld) as well as a series of ‘Coastal Belts’ (CB 1, 3,
5) which feature, similar to AT 9 Albany Coastal Belt (but unlike
the rest of the units of the National Vegetation Map), current
rather than potential vegetation patterns. We presume that
most of the CB 1, 3 and 5 were formed by subtropical forests
Elizabeth in South Africa. The southern
part of the WWF ecoregion Maputaland
Coastal Forest Mosaic (Kirkwood 2004)
and KwaZulu-Natal-Cape Coastal
Forest Mosaic (Heijnis 2004) overlap
with the IOCB as defined in this chap-
ter. The former unit is identical with our
Maputaland Coastal Belt (with all imbed-
ded intrazonal and azonal vegetation
units), while the latter extends west-
wards beyond the limits of the southern-
most vegetation unit within the IOCB (CB
5 Transkei Coastal Belt) and incorporates
also the coastal belt between the Buffels
River (East London) and Port Elizabeth.
We consider the classification of the
coastal belt of the Albany Thicket Biome
into the KwaZulu-Natal-Cape Coastal
Forest Mosaic as not justified. The latter
portion of the coast has a warm-temper-
ate character, carries climax vegetation
(Albany Thickets), various coastal veg-
etation types on the adjacent seashore,
strandveld (Vlok & Euston-Brown 2002,
Vlok et al. 2003) as well as forest vegeta-
tion (Von Maltitz et al. 2003).
The IOCB is a climatically, ecologically
and biogeographically peculiar region
that deserves standing on its own at
the level of biome within the scope of
the South African vegetation. It is the
southernmost, hence marginal, outlier of
the East African Tropical Coastal Forest
(Burgess et al. 1996, 1998, Burgess &
Clarke 2000)—a member of the global
Tropical and Subtropical Moist Forest
Biome. As such it can be classified as the
‘Subtropical Coastal Forest Biome’—an
interesting geographical analogue of the
‘Mata Atlântica’, the subtropical coastal
forest of the Atlantic seaboards of east-
ern Brazil (Hueck 1966), experiencing a
fate very similar to the IOCB (Por 1992).
The IOCB lost its natural ‘forest face’ due
Table 11.1 Indian Ocean Coastal Belt (corresponding to Subtropical Coastal Forest Biome)
as a composite of zonal, intrazonal and azonal vegetation units. For the discussion on the
concepts related to zonality, see Chapter 13 on 'Inland Azonal Vegetation'.
Vegetation Unit Extent (km2) Zonality status
FOz 7 Northern Coastal Forest 1,F 467 zonal
FOz 8 Scarp Forest 4,F 435 intrazonal (endemic)6
FOa 1 Lowveld Riverine Forest F 5 intrazonal/azonal5
FOa 2 Swamp Forest F 38 intrazonal/azonal5
FOa 3 Mangrove Forest F 33 intrazonal/azonal5
CB 4 Pondoland-Ugu Sandstone Coastal Sourveld 1297 intrazonal (endemic)7
CB 2 Maputaland Wooded Grassland 991 intrazonal (endemic)7
CB 1Maputaland Coastal Belt 4015 zonal2
CB 3 KwaZulu-Natal Coastal Belt 6293 zonal2
CB 5 Transkei Coastal Belt 1628 zonal2
AZe 3 Subtropical Estuarine Salt Marshes C 3 intrazonal/azonal5
AZs 3 Subtropical Dune Thicket C 20 intrazonal/azonal5
AZd 4 Subtropical Seashore Vegetation C 42 intrazonal/azonal5
AZf 6 Subtropical Freshwater Wetlands A 501 intrazonal/azonal5
AZa 7 Subtropical Alluvial Vegetation A 8 intrazonal/azonal5
Freshwater Lakes3 91 not applicable
Coastal Lagoons3 465 not applicable
1 incl. KwaZulu-Natal Coastal Forest and KwaZulu-Natal Dune Forest (Forest Types according
to Von Maltitz et al. 2003)
2 mapped as mosaic of primary (mainly edaphic) grassland, secondary grassland and suc-
cession thickets seral to the subtropical coastal forests (still found in patches and mapped as
Northern Coastal Forest)
3 not recognised as vegetation units (nonvegetated water bodies)
4 incl. Pondoland Scarp Forest and Eastern Scarp Forest (Forest Types according to Von
Maltitz et al. 2003)
5 intrazonal on regional scale; azonal on continental scale
6 endemism of this vegetation type relates to southern Africa (Scarp Forest occurs in Eastern
Cape, KwaZulu-Natal and Mpumalanga, Swaziland, and possibly also in Mozambique)
7 endemic to the Indian Ocean Coastal Belt
8 none mapped in Indian Ocean Coastal Belt
A discussed in Chapter 13: Inland Azonal Vegetation
C discussed in Chapter 14: Coastal Vegetation
F discussed in Chapter 12: Afrotemperate, Subtropical and Azonal Forests

572
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
(see also the modelling study by Eeley et al. 1999), and possibly
also by some more natural grassland. However, today these for-
ests and natural grasslands have been replaced by a mosaic of
secondary grasslands, seral thickets and bushveld or obliterated
by agriculture and human settlements.
2. Ecology: Climate, Geology, Soils and
Hydrology
2.1 Climate
The northern regions of the IOCB, close to the coast, have mar-
ginally nonseasonal (even) rainfall, with precipitation concen-
trated in summer. Richards Bay and St Lucia both have 41.6%
of their annual mean rainfall in winter. There are approximately
14 rain days in June and July (Camp 1999b). The winter rains
are associated with frontal systems from the south. Further
south, the rainfall seasonality of the belt, also close to the sea,
becomes clearly summer (Bailey 1979) in most places. The winter-
rainfall proportion is 35.6% for Empangeni, 30.9% for Mount
Edgecombe, 27.6% for Pinetown, 31.8% for Port Shepstone,
28.8% for Paddock, 32.7% for Cape Hermes Lighthouse (at
Port St Johns) and 36.6% at Bashee Lighthouse (near the
mouth of the Mbhashe River). The rainfall seasonality increases
gradually inland towards neighbouring savanna vegetation
types, for instance from about 33.6% on the border between
Maputaland Coastal Belt and Zululand Thornveld (Riverview)
and 26.7% on the border between KwaZulu-Natal Coastal Belt
and Ngongoni Veld (Eshowe) to 28.0% (Mkuze) and 25.1%
(Makatini Agr) well within savanna vegetation types. Strong
summer-rainfall seasonality (<20% Bailey 1979) is only found
in savanna types much further inland (e.g. in Thukela Valley
Bushveld: Muden 19.3%). On the Transkei coast Cawe (1994)
found that the proportion of winter rainfall is highest along the
coastal belt closest to the sea; he refers to this as rainfall type D
‘High rainfall with high winter rainfall’ and he found this climate
type to be confined to Acocks Veld Type 1.
In the KwaZulu-Natal part of the IOCB the mean annual rain-
fall ranges between 1 272 and 819 mm (Camp 1999b), while
on the Eastern Cape coast the mean annual rainfall reaches
1 120 mm at Cape Hermes (Port St Johns) and 1 128 mm at
Bashee Lighthouse (near the mouth of the Mbhashe River). The
east-west gradient of annual precipitation is remarkably steep,
especially in Maputaland, with around 1 200 mm on the coast,
while about 60 km inland (Ndumo in the Savanna Biome) only
about 60% of this amount occurs.
The mean annual temperature ranges relatively narrowly from
about 22°C (in the north, near the Mozambique border) to
20.4°C at Mount Edgecombe (near Durban) to 19.9°C at Cape
Hermes (Port St Johns) and to 19.1°C at the Bashee Lighthouse
(near the mouth of the Mbhashe River). Summers are hot to
very hot, while winters are mild, with hardly any frost (higher in
the southernmost parts and then only mild).
Evaporation is roughly between 1 490 mm per annum on the
South Coast (KwaZulu-Natal) and as high as 1 833 mm near
Empangeni. Air humidity is high, especially along the coast in
summer and may reach saturation point. Mean monthly relative
humidity at 14:00 at Richards Bay varies from 72% in November
to 59% in August. Moist summer heat may be a cause of dis-
comfort, but it helps to maintain the temperature-sensitive and
moisture-demanding tropical vegetation of the region—both
indigenous as well as alien (crops and ornamental flora).
The region is under the combined influence of several wind
systems (Bews 1920): rain-bearing southern and southwestern
winds, eastern afternoon sea-breezes, and autumn anticyclonic
föhns. The wind run was measured to be about 160 km per
day in Empangeni. Cyclones originating over the Indian Ocean
occasionally hit the region from the northeast, causing major
climatic and hydrological havoc, for example the 1984 cyclone
called Domoina.
The pronounced hot and damp tropical character of the climate
of the IOCB in summer and its mild and slightly drier subtropi-
cal character in winter can be ascribed to several global and
macroregional factors. Firstly, it is the unusual southbound shift
of the Intertropical Convergence Zone in summer (Tyson 1986,
Stokes et al. 1997). Secondly, it is the ameliorating influence of
the warm Agulhas Current flowing close to the eastern coasts
of South Africa. These factors control the deep intrusion of
typical tropical biota on terra firma (terrestrial flora and fauna
and associated biotic communities) on shore (global southern-
most occurrence of mangroves) and offshore (global southern-
most occurrence of corals: Ramsey 1994, Riegl 2003). Relative
humidity of at least parts of the IOCB is remarkably similar to
that in regions much further up the east coast of Africa within
the tropics. For example, mean annual relative humidity for
Durban is 79% (max. in February: 82% and min. in July: 73%)
compared to 73% (max. in April: 82% and min. in October:
67%) for Dar-es-Salaam (Tanzania) and to 74% (max. in May:
79% and min. in February: 70%) for Mombasa (Kenya) (Müller
1982). Corresponding figures for relative humidity in Beira and
especially Maputo (both in Mozambique) are considerably lower
than for Durban.
2.2 Geology and Soils
The IOCB is geologically very heterogeneous, including Mokolian-
age granites and gneisses that form the basement rocks for
Climate diagrammes
01-08-2005, 10:25, CB VB53 ColourFillReady_climate_graphs.xls
CB 1 M aputala nd Coa sta l Be lt
>2A 93 3 ^^
2A4 G 21 
>2E 21 .1 4
>75 0U
>2A 6 1904 ^ ^
>2D >D 68 
0
30
60
90
120
; 7 > 2 > ; ; 2 D @ ? 5
mm
0
10
20
30
°C
CB 2 M aputala nd W ood ed G ras slan d
>2A 96 4 ^^
2A4 G 20 
>2E 21 .0 4
>75 0U
>2A 6 1902 ^ ^
>2D >D 68 
0
30
60
90
120
; 7 > 2 > ; ; 2 D @ ? 5
mm
0
10
20
30
°C
CB 3 K waZ ulu-Natal Coast al B elt
>2A 98 9 ^^
2A4 G 20 
>2E 19 .6 4
>75 0U
>2A 6 1659 ^ ^
>2D >D 65 
0
30
60
90
120
; 7 > 2 > ; ; 2 D @ ? 5
mm
0
10
20
30
°C
CB 4 Pondo land-Ugu Sandstone Coas tal Sour veld
>2A 10 75 ^^
2A4 G 18 
>2E 18 .4 4
>75 1U
>2A 6 1549 ^ ^
>2D >D 63 
0
30
60
90
120
; 7 > 2 > ; ; 2 D @ ? 5
mm
0
10
20
30
°C
CB 5 T rans kei Coas tal Belt
>2A 10 15 ^^
2A4 G 19 
>2E 18 .4 4
>75 0U
>2A 6 1594 ^ ^
>2D >D 65 
0
30
60
90
120
; 7 > 2 > ; ; 2 D @ ? 5
mm
0
10
20
30
°C
Figure 11.2 Climate diagrams of Indian Ocean
Coastal Belt units. MAP: Mean Annual Precipitation;
APCV: Annual Precipitation Coefficient of Variation;
MAT: Mean Annual Temperature; MFD: Mean Frost
Days (days when screen temperature was below 0°C);
MAPE: Mean Annual Potential Evaporation; MASMS:
Mean Annual Soil Moisture Stress (% of days when
evaporative demand was more than double the soil
moisture supply).

STRELITZIA 19 (2006)
573
Indian Ocean Coastal Belt
sedimentary deposits spanning almost the entire Phanerozoic,
from Ordovician to Recent times, and representing all levels of
softness and the large variety of physical and chemical traits.
The most striking, and for the region most characteristic, rocks
are sandstones of the Msikaba Formation and those of the Natal
Group (Thomas et al. 1992) which support endemic vegetation
types such as Pondoland Scarp Forests (sensu Von Maltitz et al.
2003)—part of FOz 5 Scarp Forest (see Chapter 12) and CB 4
Pondoland-Ugu Sandstone Coastal Sourveld. On the Pondoland
Wild Coast these sandstones descend to sea level, where they
form dramatic coastal cliffs ornamented by numerous spec-
tacular waterfalls. In KwaZulu-Natal, most of this belt turns
inland and continues in the form of geological islands at vari-
able distance from the coast. The Msikaba Sandstones have a
high quartz content (70%–96%), with potassium feldspar and
small amounts of mica occurring as well (Hobday & Mathew
1974). They are stratigraphically linked to the Witteberg Group
of the Cape Supergroup (both are of Devonian age). The Natal
Group is, like the Peninsula Formation sandstones of the Table
Mountain Group, Ordovician. Like the sandstones building the
core of the Cape Fold Belt mountains, the soils developing over
both Msikaba and Natal Sandstones are sandy, highly leached,
acidic and mostly very shallow. Rock outcrops are very frequent
and dramatic scarps, krantzes and deep canyons are the typi-
cal accompanying geomorphological feature of this geological
landscape.
The sandstone belt (spanning Port St Johns and Port Edward)
divides the remainder of the IOCB into two distinct parts. The
northernmost region (Maputaland) is a broad coastal plain con-
sisting of a variety of Cenozoic to Recent marine sediments,
mainly of Berea and Muzi Formations. A system of dune cordons
can be traced on this plain, marking past sea-level stillstands of
various ages. The oldest dune cordon (Ndumo) is more than 25
my old and occurs outside the IOCB proper. The coastal plain
itself is of Plio-Pleistocene age (3 mya to 10 000 yrs ago; Davis
1976, Botha 1997) and is covered by sandy deposits, forming
a high dune cordon all along the coast. Between Richards Bay
and Umlalazi these dunes cover the interesting Port Dunford
Formation (Oschadleus et al. 1996), 70 000 years old. Dystric
regosols (formed by well-drained and acidic sands) are formed
on elevated slopes and crests of the dune cordons. Humic gleysols
(wet, acidic sands with abnormal accumulation of organic mat-
ter) are found in depressions—habitats with a high water table
(Matthews et al. 1999). The rest of the KwaZulu-Natal coastal
belt is a geologically complicated mosaic of Karoo Supergroup
clastic sediments, metamorphic rocks of the Namibian Erathem
and Cenozoic to Recent sediments.
Most of the Transkei Coastal Belt is built of Karoo sediments,
including sandstone and mudstone of the Adelaide Subgroup,
shale, mudstone and sandstone of the Ecca Group as well as
tillites of the Dwyka Group; Jurassic dolerite intrusions occur
in places.
Right at the coast, sand dune cordons have been formed or are
being formed—these are calcareous young Quaternary sands
derived from rocks of Cretaceous and Cenozoic origin. The sand
dunes are massive and very high, especially in Maputaland as
well as on the Zululand coast. They become rare in the southern
part of the IOBC, where they are concentrated into small pockets
at river mouths cutting though the steep coasts of Transkei.
2.3 Hydrology
The vegetation patterns of the IOCB are the result of intricate
palaeofloral history, climate, geology, and last, but not least,
hydrology. There are a number of azonal (or intrazonal) vegeta-
tion types controlled by temporal or semipermanent flooding
embedded within the IOCB (Table 1). These include freshwa-
ter wetlands, estuaries, coastal lakes, mangroves, riverine and
swamp forests, described in separate chapters on Inland Azonal
Vegetation, Coastal Vegetation and Forests.
Maputaland is a particularly water-rich environment with, for
example, spectacular coastal lagoons such as Kosi Bay and St
Lucia, big lakes (Lake Sibaya is the largest freshwater lake in
South Africa; Wright et al. 2000) and extensive swamps (Muzi
Swamp, swamps of the Mkuze floodplain north of the St Lucia
lagoon). In many places the ground-water table is very high. It
feeds the marshes and pans and is regularly replenished by rain.
Sandy substrates result in considerable fluctuations in ground-
water level (Matthews et al. 1999). The existence of some of the
grassland types, including the enigmatic Maputaland Wooded
Grasslands and other ‘edaphic’ grasslands embedded within
the Maputaland Coastal Belt and KwaZulu-Natal Coastal Belt,
is linked to dynamics of the ground-water table and temporary
surface flooding (e.g. Matthews et al. 1999).
3. Biogeographical Patterns: Past and
Present
3.1 Palaeo-ecological Patterns
According to Maud (in Tinley 1985: 43), at least the Maputaland
region must have been submerged approximately 10 mya as
witnessed by Miocene and Pliocene sediments reaching up
to 300 m above present sea level. Mid- and Upper Pliocene
were characterised by uplift and tilting of the southern African
subcontinent, resulting in formerly submerged areas (or some
parts of them) becoming dry land. However, the areas that we
define as parts of the current IOCB were formed only much
later—through a series of Pleistocene marine regressions.
It is clear that the climate in southern Africa underwent dra-
matic changes during the Pliocene and Pleistocene (past
5 my) marked by about 21 climatic cycles (each approximately
100 000 yrs long) of alternation of dry/cold and wet/warm
climates (Deacon 1983, Tyson 1986, Deacon & Lancaster 1988).
Undoubtedly these changes shaped the face of South African
vegetation and are supposed to have had major impacts on
the vegetation, especially on the extremes of macroclimatic
gradients. There is only scanty palaeo-ecological (palynologi-
cal) information on the vegetation of the IOCB during the Last
Interglacial (approximately 130 000–40 000 BP). Studies of a
lignite layer embedded in the Port Dunford Formation between
Umlalazi Lagoon near Mtunzini and Richards Bay (Scott et al.
1992, Oschadleus et al. 1996) suggest existence of a complex
of palustrine vegetation probably surrounded by coastal fresh-
water lakes, and forest vegetation in which Podocarpus (most
likely Afrocarpus falcatus) might have been dominant. These
papers hypothesised that the yellowwood forest developed
here after formation of the peat (turning lignite) layer, coincid-
ing with the deposition of sand with lower organic content.
An abundance of Podocarpus pollen was found also in the
sediments of the Lake Teza (near Mtubatuba)—around 3 400
yrs BP. Podocarpus contributed up to 20% of pollen found in
the studied profile (Scott & Steenkamp 1996) deposited since
the early Holocene. Scott & Steenkamp (1996) further imply
that the Podocarpus pollen came from an ‘original coastal
woodland environment’. Unlike Port Dunford, in the surround-
ings of Lake Teza, Podocarpus was replaced by local swampy
elements. Whatever the direction of the replacement, both
studied localities suggest a spatial and temporal link between
swamp vegetation and (presumably) Podocarpus-dominated

574
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
vegetation. Incidentally, it is the Swamp Forest, which supports
(probably relictual) populations of afrotemperate tree elements
such as Apodytes dimidiata, Ilex mitis, Rapanea melanophloeos,
Peddiea africana, Afrocarpus falcatus, Psychotria capensis and
Scutia myrtina (Wessels 1991). Afrocarpus falcatus occurs in
the IOCB not only in the Scarp Forest (which shows a number
of transitional traits between coastal and afrotemperate forests;
see Chapter 12), but also in subtropical Coastal Forest (Lubbe
1997, Van Wyk & Smith 2001). Mazus (2000) added some more
palynological data from the KwaZulu-Natal coastal peatlands
and confirmed that Podocarpus has been abundant, especially
in times when the regional climate during the Last Glacial was
presumably wetter (70 000–34 000 BP and 14 000–5 000 BP).
Botha et al. (1992) found Podocarpus pollen to be abundant
in buried paleosols from cooler and wetter periods of the Late
Pleistocene Hypothermal, but their locality is situated almost
1 000 m higher than the coastal peat deposits. Does the abun-
dant occurrence (and dominance at the Port Dunford site) of
Podocarpus in the pollen spectra mean (as argued for instance
by Mazus 2000) that the coastal belt between Richards Bay and
Umlalazi (and even further north as far as the Muzi Swamp)
supported forests of afrotemperate nature? Not necessarily, as
the tree species known today as ‘afrotemperate’ (including both
genera Podocarpus and Afrocarpus) might rather be relicts of
temporary migrations (or dispersal) from afromontane localities
during wetter periods to palustrine coastal forests. The afro-
temperate elements found within the IOCB today are exclusive
to the Swamp Forests (Venter 1972, Lubbe 1997). These, as we
know them today, ‘(sub)tropical’ Swamp Forests might have
acquired a more ‘afrotemperate’ face due to the increase of
cover of Podocarpus/Afrocarpus in these habitats during cooler
(but sufficiently wet) periods.
The vegetation landscapes of the IOCB, in concert with other
biomes of southern Africa, underwent major changes as a
response to a drop in temperature, reaching its minimum at
about 18 000 yrs BP, at Late Glacial Maximum (LGM). This
period is known to be not only cold (on average 5–6°C lower
than today, Botha et al. 1992), but also drier—with precipitation
as low as 40–70% of the present mean. The Indian Ocean sea-
boards of South Africa were much cooler and drier than today,
owing also to the lower sea-surface temperatures (Van Zinderen
Bakker 1982) and weaker and shallower Agulhas Current (Prell
et al. 1980). Low precipitation and concentration of water in ice
shields can cause a decrease of sea level by about 120 m. At
the Indian Ocean seaboards, this resulted in reworking of the
coastal dune cordons.
Following the LGM, climate ameliorated rather rapidly and wet-
ter conditions re-established in the IOCB, sometimes between
17 000–15 000 BP (Tyson 1986). Along the time axis towards
the present time, the temperature kept rising and around 7 000
BP it reached its alti-thermal (Tyson 1986, Deacon & Lancaster
1988, Partridge et al. 1990).
A new dune cordon was built and formed a corridor allow-
ing (sub)tropical flora to migrate southwards together with the
southward shift of the Intertropical Convergence Zone. It was
supposedly during this period (between 15 000 to 7 000 BP)
when the subtropical woody vegetation of the IOCB (as we
know it today) staged its come-back or re-established here. In
the process, the IOCB—the youngest biome in South Africa—
was born.
3.2 Current Biogeographical Patterns
Biogeographically the IOCB region (and, unfortunately, also
much of its deep hinterland) has been classified as Tongaland-
Pondoland Regional Mosaic, spanning the coastal regions
between Port Elizabeth and Xai Xai at the Limpopo River mouth
in Mozambique and including regions such as the former
Albany District, coastal hinterland as deep as Pietermaritzburg
and Nongoma, most of Swaziland and the southern Lowveld
(Moll & White 1978, White 1983). In a decisively altered version
this region was redefined by Van Wyk (1994; see also Van Wyk
& Smith 2001), setting the southernmost border at the Buffels
River (albeit with reservations) and by including coastal plains of
the Eastern Cape as far as Queenstown and piedmonts of the
Southern Berg and further north all regions as far as the Low
Drakensberg. Van Wyk (1994) renamed this unit to become the
‘Maputaland-Pondoland Region’, which is claimed to encom-
pass two Centres of Endemism (CE), namely the Pondoland CE
and Maputaland CE. Clarke (1998) reclassified the Maputaland
CE by incorporating it into his Swahili-Maputaland Regional
Transition Zone.
The extent of the Pondoland CE corresponds well with the core
of our vegetation unit CB 4 Pondoland-Ugu Sandstone Coastal
Sourveld. The Maputaland CE is heterogeneous in terms of
vegetation (as well as in terms of age of incorporated vegeta-
tion units). It incorporates the CB 1 Maputaland Coastal Belt,
CB 2 Maputaland Wooded Grasslands as well as part of the
accompanying forest types (see Chapter 12) and azonal wet-
land and coastal vegetation. Unlike Van Wyk & Smith (2001),
we do not recognise the Lebombo Mountains as part of the
Maputaland CE.
The IOCB appears to be the youngest biome in our region
and still two regional centres of endemism (Pondoland and
Maputaland) coincide with the extent of the IOCB. How does
this fit with the relatively ‘recent’ dramatic climatic (and palaeo-
ecological) scenarios that led to formation of this new biome?
We suggest that the explanation lies in the azonality (or intra-
zonality) of the habitats supporting endemic flora. The endemic-
rich coastal sourveld is limited to special geology (sandstones
of Msikaba Formation) supporting nutrient-stressed soils, and
in a way simulating the geological and pedological conditions
typical of the Fynbos (Van Wyk 1989, 1990a, 1994, Van Wyk &
Smith 2001). The evolutionary old endemics of the Pondoland
are concentrated either to scarp (and deep-gorge) forests,
which undoubtedly must have undergone major shrinkage dur-
ing the LGM (as well as during the much earlier Pleistocene
glacials), but were able to recover from local, well-sheltered
refugia. These forests, too, are confined largely to the same
types of sandstones and to a lesser extent also to granites
(Ongoye Forest near Empangeni). The endemics of Maputaland
are largely neo-endemics—the infraspecific status of many of
them suggests a very recent origin (Van Wyk & Smith 2001).
Within the IOCB proper, they are almost exclusively found in
azonal vegetation types, such as CB 2 Maputaland Wooded
Grassland. The occurrence of palaeo-endemics in the region
(such as Helichrysopsis septentrionale in wooded grasslands,
Encephalartos ferox in coastal thicket and coastal forest, and
palustrine palm Raphia australis) demonstrates the ability of
these taxa to resist dramatic climate changes. Resilience of hab-
itats such as wooded grasslands and swamps might have also
played a role in preserving these old relics.
The nature of regional endemism in the IOCB and the major
phytogeographic links are discussed in detail by Moll & White
(1978). These authors have established that the flora of the IOCB
contains Zanzibar-Inhambane, Zambezi, Afromontane, Upland,
Cape/Afromontane, Karoo-Namib and Guineo-Congolian link-
ing elements as well as a separate category of ‘chorological and
ecological transgressors’ occurring in a wide range of vegeta-
tion types in southern Africa.

STRELITZIA 19 (2006)
575
Indian Ocean Coastal Belt
4. Present Status
Land use is primarily sugarcane farming (over 88% of the cul-
tivated area; Camp 1999b) in the KwaZulu-Natal area of the
biome and subsistence farming in the Eastern Cape. Subsistence
farming areas from Port Shepstone in southern KwaZulu-Natal
to the KwaZulu-Natal/Mozambique border are undergoing
rapid development to small-scale sugarcane farming and small-
scale commercial tree farming. This is resulting in the loss of
vast areas of natural vegetation.
The coastline of KwaZulu-Natal south of St Lucia Estuary is highly
developed, with only small isolated occurrences of natural veg-
etation. A considerable area of the Forest Biome is embedded
within this geographical area. In contrast, the coastline of the
Transkei is little transformed and there are many natural grass-
land patches outside formal nature reserves. The IOCB supports
the highest human population concentrations on the eastern
seaboard.
About 39% of the IOCB's geographical area has been trans-
formed. However, there is a considerable range in the levels
of transformation within and between vegetation types, with
corresponding implications for meeting conservation targets for
biodiversity. South of the KwaZulu-Natal/Eastern Cape border
there remain extensive areas of natural vegetation and con-
sequently conservation targets are likely to be attainable, but
throughout the KwaZulu-Natal portion of the biome critically
high levels of transformation make it very difficult to meet con-
servation targets.
Approximately 7% of the IOCB is formally protected in statutory
reserves. However, this is disproportionately spread between the
five vegetation units, one of them with less than 1% protected.
5. Threats to Natural Vegetation
Cultivation and afforestation are the greatest threats. Extensive
areas, in many cases previously grazed on a subsistence basis,
are undergoing rapid development to small-scale sugarcane
farming and small-scale commercial tree farming. Outgrow
projects of large companies coupled with large-scale govern-
ment water supply schemes and other agricultural incentive
schemes are promoting small-scale commercial farming in areas
hitherto not available for these land uses.
Alien invasive plants are a major and growing threat.
Replacement of natural plant communities to ones dominated
by alien plants is prevalent throughout the geographical area
of the biome and the vegetation units. Chromolaena odorata is
the main problem plant. Severe loss of browsing and grazing for
both domestic livestock and wildlife is commonplace. Where
landowners can afford it, large amounts of money are often
allocated to control of invasive plants. However, in poor areas
infestation is usually so severe that little indigenous vegetation
survives.
Extensive areas of subsistence farming often occur in under-
developed areas between commercial agriculture. Grasslands
are often burnt indiscriminately to the disadvantage of many
natural plant communities and consequently of wildlife in gen-
eral. This also tends to affect the ecological functioning of bush-
clump and forest margins. Traditional hunting is sometimes the
cause of such fires.
Urbanisation is rapidly expanding into the few natural areas
remaining near the many development nodes. Some of the
most prolific examples are to be seen at Richards Bay, Durban,
Scottburgh, Port Shepstone and Margate.
The Pondoland coastal area faces the threat of dune mining and
the construction of a new N2 toll road.
6. Action: Conservation and
Management of Resources
In terms of the conditions of a World Heritage Site, the Greater
St Lucia Wetland Park is run by a Wetlands Authority. The man-
agement of the wildlife has been delegated to the provincial
nature conservation agency Ezemvelo KZN Wildlife. In addition,
many natural areas of Maputaland are benefiting from the
Lubombo Spatial Development Initiative. This applies mainly to
the Savanna Biome west of the IOCB and is discussed in that
chapter. It places much of the northern part of the IOCB in the
fortunate position of being under well-developed, secure man-
agement structures.
Ezemvelo KZN Wildlife is mandated to attend to nature con-
servation needs throughout the KwaZulu-Natal part of the
IOCB. The southern half of the coastal belt is, however, poorly
provided for in terms of well-integrated land use planning. To
this end, Ezemvelo KZN Wildlife has developed detailed con-
servation planning protocols based on systematic conservation
planning approaches. This includes the modelling of irreplace-
able areas. These are then embedded in the Municipal Spatial
Development Frameworks. Other land use planning controls
that are important for the conservation of natural vegeta-
tion include the Durban Metro Municipal Open Space System
(D-MOSS).
In the Eastern Cape a project integrating conservation and
development on the Wild Coast (most of the Transkei coast-
line and incorporating most of Pondoland), has been launched
by the Eastern Cape Department of Economic, Environmental
Affairs and Tourism and the Wilderness Foundation (The Herald
News 29/7/2004). The project will build on the 1997 Spatial
Development Initiative and is being modelled on the STEP
Programme for the Thicket vegetation type. The Wild Coast
Conservation and Development Project has initiated actions
that include conservation assessment, strategic environmental
assessment, an integrated land use plan that nests biodiversity
conservation objectives into the regional sustainable develop-
ment framework, a conservation strategy and action plan, par-
ticipation in the Global Environmental Facility (GEF) funding
proposals, and implementation programme for the Wild Coast
in conjunction with the national Department of Environmental
Affairs and Tourism. The long-term goal of the GEF project is:
Representative system of protected areas in priority bioregions
is established effectively, managed and contributes to sustain-
able development. The GEF project objective is: An effective
network of protected areas is established on the Wild Coast
and provides tested co-management models for replication.
The possible creation of a Pondoland Park will form part of this
planning.
7. Further Research Challenges
Maputaland became a focus of interest of geologists (Botha
1997, Maud & Botha 2000, Wright et al. 2000) as a prime
example of evolution of coastal plains, of vegetation ecologists
owing to interesting sources of palaeo-ecological data (Scott et
al. 1992, Scott & Steenkamp 1996, Mazus 2000), of biogeog-
raphers because of the key importance of the Belt for south-
bound plant and animal migrations (White 1983, Lawes 1990)
and of vegetation ecologists for the diversity of vegetation
types reflecting intricate soil and hydrological patterns as well

576
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
as regional climatic gradients (Moll & White 1978, Moll 1978,
1980, Lubbe 1997). The Maputaland coast, and to some extent
also southern stretches of the KwaZulu-Natal coast, with its
mosaic of grasslands, thickets and coastal forests, faces human
pressures ranging from over-development due to increasing
tourism, urban sprawl to coastal mining. These pressures have
generated considerable interest in biological research (Weisser
1978, 1987, Weisser & Marques 1979, Ward 1980). Our knowl-
edge of floristic treasures of Pondoland is increasing (Van Wyk
1990b, Van Wyk & Smith 2001), and the region continues to
yield new surprises, such as the discovery of a spectacular new
Clivia species (Murray et al. 2004). There is still no comprehen-
sive vegetation monograph for either the Pondoland coastal
sourveld or of the famous Pondoland subtropical forests. Within
the IOCB, the biota of the remainder of the Transkei coast is
poorly known. More research in all aspects of ecology should
be initiated here, especially in the light of the increased interest
of developers targeting this coastal stretch.
8. Descriptions of Vegetation Units
CB 1 Maputaland Coastal Belt
VT 1 Coastal Forest and Thornveld (97%) (Acocks 1953). LR 23 Coastal
Bushveld-Grassland (86%) (Low & Rebelo 1996). BRG 1 Moist Coast Forest
Thorn & Palm Veld (59%) (Camp 1999a, b). Coast Grassveld p.p. & Palm
Veld p.p. (Moll 1978, 1980).
Distribution KwaZulu-Natal Province (and continuing also in
southern Mozambique): Up to 35 km broad strip along the
coast of the Indian Ocean stretching from the Mozambique
border in the north to Mtunzini in the south. Altitude varies
from about 20–120 m.
Vegetation & Landscape Features Flat coastal plain origi-
nally probably densely forested in places with a wide range of
interspersed nonforest plant communities including dry grass-
lands (which include palm veld where special conditions prevail),
hygrophilous grasslands and thicket groups. Today the vegeta-
tion landscape is composed of pockets of various forest types
(separated into different vegetation units), thickets, primary and
secondary grasslands, extensive timber plantations and cane
fields. The belt of the IOCB immediately inland (only a few
kilometres wide) and parallel to the line of Northern Coastal
Forest has a characteristic appearance of very irregular dunes
with generally open vegetation and Syzygium cordatum dot-
ted prominently on the dunes, with many irregular dune slacks
interspersed. There is little to suggest that this part of the veg-
etation, e.g. between Lake Sibaya and Kosi Lake, is secondary.
The peculiar CB 2 Maputaland Wooded Grassland—still another
vegetation unit embedded within the geographical extent of
the Maputaland Coastal Belt—is treated as a separate vegeta-
tion unit (see below).
Geology & Soils Up to about 18 000 yrs old Quaternary sedi-
ments of marine origin—mainly yellowish and argillaceous redis-
tributed sands (Berea and Muzi Formations of the Maputaland
Group, respectively). Soils nutritionally very poor and well
leached, except in the interdune depressions where organic-rich
soils are sometimes found. The dominant land types include Hb
and Ha, with some contribution of Db land type.
Climate Weak rainfall seasonality near the coast tending toward
summer rainfall towards the interior. Relatively high precipita-
tion attaining annual values up to 1 200 mm in coastal localities,
decreasing rapidly to the interior. High humidity and tempera-
ture. Mean maximum and minimum monthly temperature for
Lake St Lucia Research Centre are 35.3°C and 5.5°C (for January
and June, respectively). No incidence of frost. See also climate
diagram for CB 1 Maputaland Coastal Belt (Figure 11.2).
Important Taxa Low Shrubs: Agathisanthemum bojeri (d),
Helichrysum kraussii (d), Tephrosia longipes. Small Trees & Tall
Shrubs: Syzygium cordatum (d), Acacia natalitia, Annona sene-
galensis, Apodytes dimidiata, Bridelia cathartica, Canthium
inerme, Chrysanthemoides monilifera subsp. rotundata, Euclea
natalensis subsp. natalensis, Ficus burtt-davyi, Kraussia flori-
bunda, Phoenix reclinata, Rhus natalensis, Sclerocroton inte-
gerrimum, Strychnos spinosa. Woody Climbers: Abrus precato-
rius subsp. africanus, Smilax anceps. Herbs: Achyranthes aspera,
Centella asiatica, Chamaecrista plumosa, Hermbstaedtia odo-
rata var. aurantiaca, Vernonia centaureoides, V. oligocephala.
Graminoids: Diheteropogon amplectens (d), Eragrostis scle-
rantha (d), Ischaemum fasciculatum (d), Themeda triandra
(d), Urelytrum agropyroides (d), Aristida stipitata subsp. gra-
ciliflora, Cymbopogon pospischilii, Elionurus muticus, Eragrostis
Figure 11.3 CB 1 Maputaland Coastal Belt: Seasonally wet grasslands of the Palm Veld (with
Hyphaene coriacea) near KaNgwanase in Maputaland, northern KwaZulu-Natal.
inamoena, E. lappula, Sporobolus subu-
latus, Trachypogon spicatus, Trichoneura
grandiglumis, Tristachya leucothrix.
Biogeographically Important Taxa
(CCoastal belt element, FGeneric fynbos
element, ILPIsolated lowland populations,
MMaputaland endemic, NNorthern distri-
bution limit, SSouthern distribution limit)
Geoxylic Suffrutex: Diospyros galpiniiS.
Low Shrubs: Indigofera williamsoniiC, Rhus
kwazuluanaM, Stylosanthes fruticosaS.
Small Trees & Tall Shrubs: Hyphaene coria-
ceaS (d), Ozoroa obovataS, Rhus nebulosaC,
Synaptolepis kirkiiM. Woody Climber:
Dalbergia obovataC. Herbs: Helichrysopsis
septentrionaleM, Helichrysum tongenseC,
H. cymosum subsp. cymosumN, Nidorella
tongensisM, Senecio ngoyanusC, Vernonia
natalensisILP
. Megaherb: Strelitzia nico-
laiC (d). Succulent Herb: Orbea longi-
densM. Semiparasitic Herb: Striga
junodiiS. Graminoid: Monocymbium
ceresiiformeILP
.
L. Mucina

STRELITZIA 19 (2006)
577
Indian Ocean Coastal Belt
Endemic Taxa (FGeneric fynbos element) Herbs: Helichrysum
adenocarpum subsp. ammophilum, Vahlia capensis subsp. vul-
garis var. longifolia. Geophytic Herbs: Asclepias gordon-grayae,
Kniphofia leucocephala, Raphionacme lucens. Graminoid:
Restio zuluensisF (d).
Conservation Vulnerable. Target 25%. 15% statutorily
conserved in the Greater St Lucia Wetland Park as well as
in Sileza, Enseleni and Amathikulu Nature Reserves. More
than 30% transformed for plantations and cultivation and
by urban sprawl. Aliens include scattered populations of
Chromolaena odorata and Lantana camara. Erosion is mostly
very low. This vegetation type has a relatively high number
of plant taxa at the southernmost and northernmost limits
of their distribution range—the occurrence of widely disjunct
or outlier populations increases the conservation value of this
vegetation type.
Remark 1 The primary grasslands of interdune depressions
and seasonally waterlogged bottomlands of the Maputaland
were classified by Matthews et al. (1999) as the Eragrostis
lappula–Helichrysopsis septentrionalis and Ischaemum fas-
ciculatum–Eragrostis inamoena hygrophilous grasslands, and
by Lubbe (1997) as Ischaemum fasciculatum–Centella asia-
tica hygrophilous grassland. Like the Maputaland Wooded
Grasslands, these primary grasslands are home to a number of
Maputaland endemics such as the enigmatic Restio zuluensis
and Helichrysopsis septentrionale.
Remark 2 Most of the Maputaland Coastal Belt is agricultural
land and very little of this unit remains in a natural state in the
South African part of Maputaland. A much larger area of well-
preserved coastal belt is found in Mozambique.
References Venter (1972), Moll (1972, 1978, 1980), Moll & White (1978),
Weisser (1978, 1987), Weisser & Marques (1979), Lubbe (1997), Camp
(1999a, b), Matthews et al. (1999), Smith (2001), Van Wyk & Smith (2001).
CB 2 Maputaland Wooded Grassland
VT 1 Coastal Forest and Thornveld (100%) (Acocks 1953). LR 23 Coastal
Bushveld–Grassland (98%) (Low & Rebelo 1996). BRG 1 Moist Coast Forest
Thorn & Palm Veld (61%) (Camp 1999a, b). Incl. Themedo–Salacietum Myre
(1964).
relatively high-lying level plains. Water table found at depth 1.6–
2.0 m below surface (and slightly deeper) in average rainfall years.
Ha is the overwhelmingly dominant land type, followed by Hb
land type.
Climate Approximately the same as for the CB 1 Maputaland
Coastal Belt (both units form an intricate regional mosaic). See
also climate diagram for CB 2 Maputaland Wooded Grassland
(Figure 11.2).
Important Taxa (#Suffrutex form) Geoxylic Suffrutices: Parinari
curatellifolia (d), Salacia kraussii (d), Ancylobotrys petersiana,
Diospyros galpinii, Eugenia capensis#, Syzygium cordatum#.
Graminoids: Diheteropogon amplectens (d), Themeda tri-
andra (d), Aristida stipitata subsp. graciliflora, Bewsia biflora,
Cyperus obtusiflorus, C. tenax, Digitaria natalensis, Eustachya
paspaloides, Setaria sphacelata, Sporobolus fimbriatus, S. subu-
latus, Urelytrum agropyroides. Herb: Chamaecrista plumosa.
Geophytic Herb: Cyrtanthus galpinii. Low Shrubs: Helichrysum
kraussii (d), Agathisanthemum bojeri, Crotalaria monteiroi var.
monteiroi. Small Trees & Tall Shrubs: Acridocarpus natalitius var.
linearifolius, Dichrostachys cinerea subsp. nyassana, Diospyros
lycioides subsp. sericea, Hyphaene coriacea, Terminalia sericea.
Biogeographically Important Taxa (CCoastal belt element,
MMaputaland endemic, SSouthern distribution limit) Geoxylic
Suffrutices: Eugenia albanensisC, Gymnosporia markwardiiM;
Graminoids: Abildgaardia hygrophilaC, Cyperus natalensisC.
Herbs: Helichrysopsis septentrionaleM; Oxygonum robustumM,
Tricliceras mossambicenseM. Tall Shrub: Grewia microthyrsaS.
Woody Climbers: Albertisia delagoensisS, Cissampelos hirtaS.
Endemic Taxa (#Suffrutex form) Geoxylic Suffrutices: Ochna sp.
nov., Syzygium cordatum#. Succulent Herb: Aloe sp. nov. (Strey
5100 PRE). Geophytic Herb: Brachystelma vahrmeijeri.
Conservation Endangered. Target 25%. About 17% statutorily
conserved mainly in the Greater St Lucia Wetland Park. Some
46% transformed mostly for plantations and partly for culti-
vated land. The southern half of the area is not protected and
it is here that over 90% of the extent of the vegetation type
has been transformed—mostly to pulpwood timber plantations,
cane fields and informal settlements. Aliens include scattered
populations of Chromolaena odorata and Lantana camara.
W.S. Matthews
Figure 11.4 CB 2 Maputaland Wooded Grassland: Wooded grassland in Maputaland (north-
ern KwaZulu-Natal) with prominent (silvery leaves) undescribed species of geoxylic suffrutex
(Ozoroa sp. nov.).
Distribution KwaZulu-Natal Province
and southern Mozambique: In South
Africa from the Mozambique border near
KwaNgwanase southwards to Sileza,
Sibaya, Mseleni, Mbazwana, Sodwana
Bay, Ozabeni, eastern and western shores
of Lake St Lucia, KwaMbonambi and as
far south as near Richards Bay. Altitude
varies from about 20–120 m.
Vegetation & Landscape Features
Generally flat landscape of the
Maputaland coastal plain supporting
coastal sandy grasslands rich in geoxy-
lic suffrutices, dwarf shrubs, small trees
and very rich herbaceous flora. Excluded
from this unit are the many interdune
depression wetlands and hygrophilous
grasslands neighbouring the wooded
grasslands.
Geology & Soils Quaternary redistrib-
uted sand supporting yellowish redis-
tributed sands of the Berea Formation
(Maputaland Group). These are dystric
regosols building dune crests, slopes and

578
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
present the KwaZulu-Natal Coastal Belt
is affected by an intricate mosaic of very
extensive sugarcane fields, timber plan-
tations and coastal holiday resorts, with
interspersed secondary Aristida grass-
lands, thickets and patches of coastal
thornveld.
Geology & Soils Ordovician Natal Group
sandstone, Dwyka tillite, Ecca shale and
Mapumulo gneiss (Mokolian) dominate
the landscapes of the KwaZulu-Natal
Coastal Belt. Weathering of old dunes
has produced the red sand, called the
Berea Red Sand, in places. The soils sup-
ported by the above-mentioned rocks are
shallow over hard sandstones and deeper
over younger, softer rocks. Fa land type
dominates the area, while Ab land type
is only of minor importance.
Climate Summer rainfall, but with some
rainfall also in winter. High air humidity.
No incidence of frost. Mean maximum
Remarks This type is an example of the famous ‘underground
forests of Africa’ (White 1976) characterised by plants with
sometimes enormous underground woody parts connect-
ing apparently separate dwarf shrubs or even with only tufts
of leaves above ground. This growth form is called a geoxylic
suffrutex (dwarf woody plant with annual or short-lived
above-ground woody shoots sprouting from massive under-
ground ‘stem’) (White 1976, Matthews et al. 1999, Van Wyk &
Smith 2001). Some of the taxa occur naturally only as geoxylic
suffrutices, while in some, generally more widely distributed
taxa (incl. Eugenia capensis, Syzygium cordatum) suffrutex
forms occur in these habitats. Several hypotheses have been
suggested to explain the existence of this enigmatic vegetation
type as well as the peculiar concentration of the geoxylic suffru-
tices. Fire-resistance and adaptation to high groundwater tables
have been proposed most often (see White 1976, Matthews
et al. 1999), but none of these (and other) explanations have
been accompanied by conclusive evidence. Species with the
geoxylic suffrutex form are also found within many open and
well-wooded savanna types including some with sandy sub-
strate where there is commonly a higher investment in below-
ground organs.
References Myre (1964, 1971), White (1976, 1983), Moll (1978, 1980),
Moll & White (1978), Van Wyk (1994, 1996), Lubbe (1997), Mathews et al.
(1999), Smith (2001), Van Wyk & Smith (2001), Felton (2002).
CB 3 KwaZulu-Natal Coastal Belt
VT 1 Coastal Forest and Thornveld (81%) (Acocks 1953). LR 23 Coastal
Bushveld–Grassland (62%) (Low & Rebelo 1996). BRG 1 Moist Coast Forest
Thorn & Palm Veld (89%) (Camp 1999a, b).
Distribution KwaZulu-Natal Province: Long and in places broad
coastal strip along the KwaZulu-Natal coast, from near Mtunzini
in the north, via Durban to Margate and just short of Port
Edward in the south. Altitude ranges from about 20–450 m.
Vegetation & Landscape Features Highly dissected undu-
lating coastal plains which presumably used to be covered to
a great extent with various types of subtropical coastal forest
(the remnants of one of which are described in Chapter 12 as
Northern Coastal Forest). Some primary grassland dominated by
Themeda triandra still occurs in hilly, high-rainfall areas where
pressure from natural fire and grazing regimes prevailed. At
Figure 11.5 CB 3 KwaZulu-Natal Coastal Belt: Complex of primary species-rich grasslands and
subtropical forests in Vernon Crookes Nature Reserve near Umzinto, KwaZulu-Natal.
and minimum monthly temperatures for Durban (airport) are
32.6°C and 5.8C and for Port Shepstone 30.6°C and 8.8°C
(both for January and July, respectively). See also climate dia-
gram for CB 3 KwaZulu-Natal Coastal Belt (Figure 11.2).
Important Taxa Graminoids: Aristida junciformis subsp. galpinii
(d), Digitaria eriantha (d), Panicum maximum (d), Themeda trian-
dra (d), Alloteropsis semialata subsp. eckloniana, Cymbopogon
caesius, C. nardus, Eragrostis curvula, Eulalia villosa, Hyparrhenia
filipendula, Melinis repens. Herbs: Berkheya speciosa subsp.
speciosa (d), Cyanotis speciosa (d), Senecio glaberrimus (d),
Alepidea longifolia, Centella glabrata, Cephalaria oblongifolia,
Chamaecrista mimosoides, Conostomium natalense, Crotalaria
lanceolata, Dissotis canescens, Eriosema squarrosum, Gerbera
ambigua, Hebenstretia comosa, Helichrysum cymosum subsp.
cymosum, H. pallidum, Hibiscus pedunculatus, Hybanthus
capensis, Indigofera hilaris, Pentanisia prunelloides subsp.
latifolia, Senecio albanensis, S. bupleuroides, S. coronatus, S.
rhyncholaenus, Sisyranthus imberbis, Stachys aethiopica, S.
nigricans, Vernonia galpinii, V. oligocephala. Geophytic Herbs:
Bulbine asphodeloides, Disa polygonoides, Hypoxis filiformis,
Ledebouria floribunda, Pachycarpus asperifolius, Schizocarphus
nervosus, Tritonia disticha. Low Shrubs: Clutia pulchella, Gnidia
kraussiana, Phyllanthus glaucophyllus, Tephrosia polystachya.
Woody Climbers: Abrus laevigatus, Asparagus racemosus,
Smilax anceps. Small Trees & Tall Shrubs: Bridelia micrantha (d),
Phoenix reclinata (d), Syzygium cordatum (d), Acacia natalitia,
Albizia adianthifolia, Antidesma venosum.
Biogeographically Important Taxa (CCoastal belt element,
SSouthern distribution limit) Graminoids: Cyperus natalen-
sisC, Eragrostis lappulaS. Herbs: Helichrysum longifoliumC,
Selago tarachodesC, Senecio dregeanusC, Sphenostylis angus-
tifoliaS. Geophytic Herbs: Kniphofia gracilisC, K. littoralisC,
K. rooperiC, Pachystigma venosumS, Zeuxine africanaS. Low
Shrubs: Helichrysum kraussiiS (d), Agathisanthemum bojeriS,
Desmodium dregeanumC. Megaherb: Strelitzia nicolaiC (d).
Geoxylic Suffrutices: Ancylobotrys petersianaS, Eugenia alban-
ensisC, Salacia kraussiiS. Small Trees & Tall Shrubs: Anastrabe
integerrimaC (d), Acacia nilotica subsp. kraussianaS.
Endemic Taxa Herb: Vernonia africana (extinct). Geophytic Herb:
Kniphofia pauciflora. Low Shrub: Barleria natalensis (extinct).
Conservation Endangered. Target 25%. Only very small part
statutorily conserved in Ngoye, Mbumbazi and Vernon Crookes
L. Mucina

STRELITZIA 19 (2006)
579
Indian Ocean Coastal Belt
Nature Reserves. About 50% transformed for cultivation, by
urban sprawl and for road-building. Aliens include Chromolaena
odorata, Lantana camara, Melia azedarach and Solanum mauri-
tianum. Erosion is low and moderate.
References Edwards (1967), Moll & White (1978), Ward (1980), Roberts
(1993), Camp (1999a, b), Heijnis (2004), Van der Linden et al. (2005).
CB 4 Pondoland-Ugu Sandstone Coastal
Sourveld
VT 1 Coastal Forest and Thornveld (53%) (Acocks 1953). LR 48 Coastal
Grassland (38%), LR 42 Moist Upland Grassland (28%) (Low & Rebelo 1996).
BRG 1 Moist Coast Forest Thorn & Palm Veld p.p. Camp (1999a, b).
Distribution Eastern Cape and KwaZulu-Natal Provinces:
Elevated coastal sandstone plateaus from Port St Johns on the
Pondoland coast (Eastern Cape) to the vicinity of Port Shepstone
(Ugu District, KwaZulu-Natal), incl. the sourveld of the well-
known Oribi Gorge. Altitude ranges from about 0–600 m.
Vegetation & Landscape Features Coastal peneplains and
partly undulating hills with flat table-lands and very steep
slopes of river gorges. These sites support natural, species-rich
grassland punctuated with scattered low shrubs or small trees
(sometimes with bush clumps, especially in small gullies). Rocky
outcrops and krantzes are common and dramatic sea-cliffs
occur. Proteaceous trees (Protea, Faurea) can be locally com-
mon where conditions allow. Although less important here, the
geoxylic suffrutex growth form (so typical of CB 2 Maputaland
Wooded Grassland), is also represented in this sourveld.
Geology & Soils This unit is strictly delimited by its geology—it
is built of hard, white, coarse-grained, siliceous quartz arenites
(sandstones) of the Msikaba Formation of the Devonian Period
(Thomas et al. 1992) giving rise to shallow, nutrient-poor (highly
leached), skeletal, acidic sandy soils. Almost 80% of the area is
classified as Fa land type, followed by Aa land type (10%).
Climate Summer rainfall with some rain in winter. No or very
infrequent incidence of frost. Mean maximum and minimum
monthly temperatures at Paddock (near Oribi Gorge in the
north) are 32.2°C and 5.8°C (for January and July, respectively).
The corresponding values for Cape Hermes Lighthouse (Port
subsp. adenocarpum, H. aureum var. monocephalum, H. her-
baceum, H. nudifolium var. pilosellum, H. pallidum, Indigofera
hilaris, Pentanisia prunelloides subsp. latifolia, Pimpinella caf-
fra, Vernonia capensis. Geophytic Herbs: Brachystelma tenel-
lum, Eriospermum mackenii. Low Shrubs: Athrixia phylicoides,
E. natalensis, E. natalitia, Gnidia anthylloides, G. kraussiana,
G. nodiflora, Leonotis intermedia, Polygala hottentotta. Small
Trees & Tall Shrubs: Euryops brevipapposus, Syzygium cordatum.
Semiparasitic Shrubs: Thesium acutissimum, T. cupressoides.
Biogeographically Important Taxa (CCoastal belt ele-
ment, EEastern isolated occurrence, FGeneric fynbos ele-
ment, NNorthern distribution limit, SSouthern distribution
limit) Geoxylic Suffutex: Gymnosporia vanwykiiC. Graminoids:
Loudetia simplexS (d), Calopsis paniculataF, Tetraria robustaEF.
Herbs: Helichrysum auricepsS, H. natalitiumS, H. pannosumS,
Senecio dregeanusS, S. rhyncholaenusS, Berkheya insignisS,
Eriosema acuminatumC, Helichrysum acutatumS, H. longifo-
liumC, Peucedanum natalenseC, Roella glomerataF,C. Geophytic
Herbs: Stenoglottis woodiiS, Asclepias patensC, Disperis woo-
diiC, Kniphofia rooperiC. Low Shrubs: Senecio medley-woodiiS,
Gnidia woodiiS (d), Agathosma ovataF, Erica aspalathifoliaC,
Gnidia coriaceaN, Muraltia lancifoliaF, Pseudarthria hookeri F,S,
Relhania pungensF, Stangeria eriopusC, Syncolostemon rotundi-
foliusC. Geoxylic Suffutex: Eriosemopsis subanisophyllaS. Small
Trees & Tall Shrubs: Faurea salignaS (d), Protea roupelliae subsp.
roupelliaeF (d), Encephalartos cafferN, Loxostylis alataF, Polygala
gazensis (isolated populations; also Inyanga), Protea caffra
subsp. caffraF, P. simplexF, Sclerocroton integerrimumS.
Endemic Taxa (FGeneric fynbos element) Graminoid: Fimbristylis
variegata. Herbs: Eriosema umtamvunense, Geranium sparsiflo-
rum, Lotononis bachmanniana, Selago peduncularis, Senecio
erubescens var. incisus, Geophytic Herbs: Brachystelma aust-
rale, B. kerzneri, Watsonia inclinataF, W. mtamvunaeF. Geoxylic
Suffrutex: Rhus acocksii. Low Shrubs: Leucadendron spissifo-
lium subsp. natalenseF (d), L. spissifolium subsp. oribinumF (d),
Acalypha sp. nov. (Scott-Shaw 636 NU), Anthospermum streyi,
Erica abbottii, E. cubica var. natalensisF, Eriosema dregei, E. lati-
folium, E. luteopetalum, Euryops leiocarpus, Gnidia triplinervis,
Leucadendron pondoenseF, Leucospermum innovansF, Raspalia
trigynaF, Struthiola pondoensisF, Syncolostemon ramulosus,
Tephrosia bachmannii. Tall Shrub: Tephrosia pondoensis.
Figure 11.6 CB 4 Pondoland-Ugu Sandstone Coastal Sourveld: Pondoland sourveld on the
edge of the Umtamvuna Gorge (near Port Edward, KwaZulu-Natal) with scattered trees of
Protea roupelliae subsp. roupelliae. The dominant grasses are Aristida junciformis and
Loudetia simplex.
St Johns, in the south) are 29.5°C and
9.6°C for the same months. See also
climate diagram for CB 4 Pondoland-
Ugu Sandstone Coastal Sourveld (Figure
11.2).
Important Taxa Graminoids:
Alloteropsis semialata subsp. eckloniana
(d), Aristida junciformis subsp. galpinii (d),
Cymbopogon nardus (d), Themeda trian-
dra (d), Tristachya leucothrix (d), Cyperus
rupestris, Diheteropogon amplectens,
Elionurus muticus, Eragrostis capensis, E.
plana, Eulalia villosa, Heteropogon con-
tortus, Panicum natalense, Trachypogon
spicatus. Herbs: Chaetacanthus burchellii
(d), Cyanotis speciosa (d), Helichrysum
allioides (d), H. appendiculatum (d), H.
krebsianum (d), H. spiralepis (d), Pentanisia
angustifolia (d), Rhynchosia totta (d),
Tephrosia macropoda (d), Berkheya
speciosa subsp. speciosa, Cephalaria
oblongifolia, Chamaecrista mimosoides,
Eriosema salignum, Euphorbia eri-
coides, Helichrysum adenocarpum
L. Mucina

580
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
Conservation Vulnerable (one of the
top six vegetation units with the high-
est level of overall vulnerability in South
Africa). Target 25%. Only about 7%
statutorily conserved in the Mkambati
Wildlife Reserve & Marine Sanctuary, and
Umtamvuna, Mbumbazi and Oribi Gorge
Nature Reserves. About 29% trans-
formed for cultivation and plantations or
by urban sprawl. In the Eastern Cape the
land use is mostly subsistence farming.
Erosion is very low and low.
Remark 1 The sandstone geology
links Pondoland to other ‘sourvelds’
of South Africa. Pondoland forms the
lowest step along a staircase of nutri-
ent-poor geologies, comprising further
the early Palaeozoic Natal Sandstones
of the KwaZulu-Natal Midlands and the
Late Triassic Clarens Sandstones of the
Drakensberg (see also Van Wyk 1994).
The occurrence of Protea roupelliae, P.
simplex, P. welwitschii, Erica natalitia,
Helichrysum herbaceum, H. krebsianum,
slopes of low-reach river valleys and coastal ridges, sometimes
broad enough to form small plains. A mosaic of grassland veg-
etation on the higher lying areas and characteristically on hill
tops and upper hill slopes, alternating with bush clumps and
small forests (considered as part of the vegetation unit FOz 5
Scarp Forest) is the major vegetation feature of the region. Most
of the grasslands are undoubtedly secondary (result of forest
clearing for cattle grazing). At the seaward border this vegeta-
tion mosaic is fringed by an interrupted belt of coastal dune
thicket (considered as part of AZs 3 Subtropical Dune Thicket)
and vegetation of young coastal habitats (dunes and beaches).
Geology & Soils Most of the area is built of Karoo Supergroup
sediments including sandstone and mudstone of the Adelaide
Subgroup, shale, mudstone and sandstone of the Ecca Group
as well as tillite of the Dwyka group. Intrusions of Jurassic Karoo
Dolerite Suite occur in places. The dominating soil forms are
Glenrosa and Mispah. Fa land type dominates the area.
Climate Summer rainfall with some rain in winter (with up to
36.6% rainfall in winter at Bashee Lighthouse). No incidence of
frost. Bashee Lighthouse recording a mean minimum tempera-
ture of 7.7°C in July. See also climate diagram for CB 5 Transkei
Coastal Belt (Figure 11.2).
Important Taxa Graminoids: Aristida junciformis subsp. galpi-
nii (d), Stenotaphrum secundatum (d), Abildgaardia ovata,
Cynodon dactylon, Dactyloctenium aegyptium, Ehrharta erecta
var. erecta, Setaria plicatilis, S. sphacelata, Sporobolus africanus.
Herb: Ipomoea cairica. Geophytic Herb: Bonatea speciosa var.
antennifera. Low Shrubs: Anisodontea scabrosa, Passerina rigida.
Succulent Herb: Crassula multicava subsp. multicava. Small Trees
& Tall Shrubs: Acacia natalitia, Cestrum laevigatum, Grewia occi-
dentalis var. occidentalis. Succulent Tree: Aloe ferox.
Biogeographically Important Taxa (all coastal belt elements)
Herb: Stachys comosa. Geophytic Herbs: Asclepias patens,
Strelitzia reginae. Geoxylic Suffrutex: Gymnosporia vanwykii.
Low Shrub: Pavetta revoluta.
Conservation Vulnerable. Target 25%. Only about 1% statu-
torily conserved, for example in Dwesa-Cwebe Wildlife Reserve
& Marine Sanctuary, Silaka Wildlife Reserve and Hluleka Wildlife
Reserve & Marine Sanctuary. About 20% transformed mainly
for cultivation. Erosion is low and moderate.
Figure 11.7 CB 5 Transkei Coastal Belt: Coastal grasslands and subtropical dune thickets near
Umgazi River Mouth on the Transkei Coast (Eastern Cape Province).
L. Mucina
H. pannosum, Senecio rhyncholaenus and Schizoglossum atro-
purpureum subsp. virens is indicative of this link. Pondoland
is a crossroads of old migration routes and perhaps also a
migration cul-de-sac (Van Wyk 1990a) of some of them. It
shows not only a clear Drakensberg link, but also clear bio-
geographical (and geological) relationships to the Capensis
through the occurrence of genera such as Agathosma, Aristea,
Athrixia, Calopsis, Cliffortia, Erica, Euryops, Leucadendron,
Leucospermum, Loxostylis, Muraltia, Phylica, Podalyria,
Prionium, Protea, Pseudoscolopia, Raspalia (the only repre-
sentative of the family Bruniaceae outside Capensis), Restio,
Relhania, Roella, Struthiola, Tetraria and Watsonia. Some of
these disjunctions occur at the species level (!): Calopsis panicu-
lata, Cliffortia odorata, Helichrysum diffusum, Loxostylis alata,
Prionium serratum, Pseudoscolopia polyantha and Restio trit-
iceus (Midgley 1986, Carbutt & Edward 2001).
Remark 2 Slight depressions on the coastal plateau and rock
pools on rocky outcrops support another suite of local endem-
ics or biogeographically important taxa linked to hygromor-
phic soils. These include Kniphofia rooperi, Podalyria velutina,
Psoralea abbottii, Utricularia sandersonii, Watsonia bachmannii
and W. pondoensis.
References Moll & White (1978), Midgley (1986), Shackleton (1989, 1992),
Shackleton (1990), Van Wyk (1990a, b), Shackleton et al. (1991), Abbott
(1993), Shackleton & Shackleton (1994), Le Roux (1995), Glen (1996), Camp
(1999a, b), Scott-Shaw (1999), Abbott et al. (2000), Carbutt & Edwards
(2001), Van Wyk & Smith (2001).
CB 5 Transkei Coastal Belt
VT 1 Coastal Forest and Thornveld (84%) (Acocks 1953). LR 48 Coastal
Grassland (37%), LR 23 Coastal Bushveld–Grassland (21%) (Low & Rebelo
1996).
Distribution Eastern Cape Province: Narrow coastal strip along
the Wild Coast of Transkei and the Indian Ocean seaboards
between Port St Johns (Egossa Interval) as far as the vicinity of
the Great Kei River in the south. Altitude ranges from about
20–450 m.
Vegetation & Landscape Features The Transkei Coastal Belt
is highly dissected, hilly coastal country with alternating steep

STRELITZIA 19 (2006)
581
Indian Ocean Coastal Belt
Remarks The nonforest vegetation of the Transkei Coastal Belt is
one of the most poorly studied vegetation types in the country.
References Acocks (1953, 1988), Moll & White (1978), Hoffman (1983).
9. Credits
The delimitation of the IOCB is based on K.G.T. Camp’s map of
Bioresource Groups for KwaZulu-Natal (Camp 1999a) within
the borders of KwaZulu-Natal and on analysis based on inter-
pretation of satellite-image data by D.B. Hoare in the Eastern
Cape Province. The borders between CB 1 and CB 3 follow,
to a great extent, the Camp’s (1999a) map, but have been
modified by C.R. Scott-Shaw and L. Mucina. The concept of
CB 2 was jointly defined by W.S. Matthews, C.R. Scott-Shaw
and L. Mucina, partly using the sources by Smith (2001) and
Felton (2002). The concept of CB 3 resulted from fusion, as
suggested by L. Mucina, and C.R. Scott-Shaw, of several units
defined by Camp (1999a). The extent of CB 1 and CB 2 was
partly defined also by mapping of the wetlands in Maputaland
by M.C. Rutherford and L.W. Powrie. The extent of all IOCB
vegetation units was also modified by the extent of the forest
patches (see Chapter 12 for Credits).
The descriptions of CB 1 to CB 4 were a joint effort by L. Mucina
and C.R. Scott Shaw; W.S. Matthews contributed to descrip-
tions of CB 1 and CB 2; L. Mucina wrote the description of CB
5. The species lists were created by L. Mucina, C.R. Scott-Shaw
and W.S. Matthews (the last-named for the Maputaland units).
The introductory text was written by L. Mucina (sections 1 to
3), while C.R. Scott-Shaw contributed sections 4 to 7 of the
introductory text. M.C. Rutherford contributed to sections 1
and 2 of the introductory text as well as to the climate and
conservation sections of the vegetation unit descriptions. Table
11.1 was created jointly by L. Mucina, M.C. Rutherford and
L.W. Powrie. The last-mentioned two authors also provided all
climate diagrams. L. Mucina (with help of C.R. Scott-Shaw) col-
lated the list of references. The photographs were contributed
by W.S. Matthews and L. Mucina. M. Rouget, and others within
the Directorate of Biodiversity Programmes, Policy & Planning
of SANBI, provided quantitative information for each vegeta-
tion unit on conservation status and targets, areas currently
conserved and areas transformed.
Ezemvelo KZN Wildlife kindly provided data on the extent of CB
2 as well as forest patches imbedded within the IOCB (Ezemvelo
KZN Wildlife 2004. Metadatabase file: KwaZulu-Natal Forest
Types, Dataset ID 550. Ezemvelo KZN Wildlife Scientific Services
Branch, Pietermaritzburg). The wetlands were mapped using
selected data from the National Land Cover 2000 project as
well as digitising by L.W. Powrie from topographic maps of a
number of wetlands in the Maputaland region. P.S. Goodman
contributed valuable comments and C. Oellerman assisted
C.R. Scott-Shaw with GIS work. R.A. Ward kindly corrected the
geological terminology.
10. References
Abbott, A. 1993. The Umtamvuna Nature Reserve. In: The Umtamvuna
Nature Reserve, pp. 2–4. Umtamvuna River Trust, Port Edward.
Abbott, A.T.D., Van Wyk, A.E., Johnson, D.N. & Scott-Shaw, C.R. 2000.
Checklist of the macrofungi, lichens, bryophytes and vascular plants of the
Umtamvuna Nature Reserve. Lammergeyer 46: 1–74.
Acocks, J.P.H. 1953. Veld types of South Africa. Mem. Bot. Surv. S. Afr. No.
28: 1–192.
Acocks, J.P.H. 1988. Veld types of South Africa, edn 3. Mem. Bot. Surv. S.
Afr. No. 57: 1–146.
Bailey, H.P. 1979. Semi-arid climates: their definitions and distribution. In:
Hall, A.E., Cannell, G.H. & Lawton, H.W. (eds), Agriculture in semi-arid
environments, pp. 73–97. Springer, Berlin.
Bews, J.W. 1920. The plant ecology of the coast belt of Natal. Ann. Natal
Mus. 4: 367–467.
Botha, G.A. 1997. Maputaland focus on the Quaternary evolution of the
south-east African coastal plain. In: Botha, G.A. (ed.), International Union
for Quaternary Research Workshop Abstracts. Council for Geosciences,
Pretoria.
Botha, G.A., Scott L., Vogel J.C. & Von Brunn, V. 1992. Palaeosols and pal-
aeoenvironments during the ‘Late Pleistocene Hypothermal’ in Northern
Natal. S. Afr J. Sci. 88: 508–512.
Burgess, N.D. & Clarke, G.P. (eds) 2000. The coastal forests of eastern Africa.
IUCN Publ., Cambridge.
Burgess, N.D., Clarke, G.P. & Rodgers, W.A. 1998. Coastal forests of eastern
Africa: status, species endemism and its possible causes. Biol. J. Linn. Soc.
64: 337–367.
Burgess, N. D’Amico Hales, J., Underwood, E., Dinerstein, E., Olson,
D., Itoua, I., Schipper, J., Ricketss, T. & Newman, K. 2004. Terrestrial ecore-
gions of Africa and Madagascar—a conservation assessment. Island Press,
Washington.
Burgess, N.D., FitzGibbon, C. & Clarke, G.P. 1996. Coastal forests of East
Africa. In: McClanaghan, T. & Young, T.P. (eds), Ecosystems and their con-
servation in East Africa, pp. 329–359. Oxford Univ. Press, Oxford.
Camp, K. 1999a. The Bioresource Units of KwaZulu-Natal. 1: 500,000.
Natural Resources, Section Technology Development and Training, KwaZulu-
Natal Department of Agriculture, Cedara. (electronic shape file of map)
Camp, K. 1999b. The Bioresource Groups of KwaZulu-Natal. Coast & Coast
Hinterland. Cedara Report No. N/A/99/12, KwaZulu-Natal Department of
Agriculture, Cedara.
Carbutt, C. & Edwards, T. 2001. Cape element on high-altitude and edaphic
islands: historical aspects and preliminary phytogeography. Syst. Geogr.
Plants 71: 1033–1061.
Cawe, S.G. 1994. Rainfall and vegetation patterns in Transkei and the envi-
rons. S. Afr. J. Sci. 90: 79–85.
Clarke, G.P. 1998. A new regional centre of endemism in Africa. In: Cutler,
D.F., Huxley, C.R. & Lock, J.M. (eds), Aspects of ecology, taxonomy and
chorology of the floras of Africa and Madagascar, pp. 53–65. Royal Botanic
Gardens, Kew.
Davis, O. 1976. The older coastal dunes in Natal and Zululand and their rela-
tion to former shorelines. Ann. S. Afr. Mus. 71: 19–32.
Deacon, H.J. 1983. Another look at the Pleistocene climates of South Africa.
S. Afr. J. Sci. 79: 325–328.
Deacon, J. & Lancaster, N. 1988. Late Quaternary palaeoenvironments of
southern Africa. Clarendon Press, Oxford.
Edwards, D. 1967. A plant ecological survey of the Tugela River basin. Town
and Regional Planning Commission, Pietermaritzburg.
Eeley, H.A.C., Lawes, M.J. & Piper, S.E. 1999. The influence of climate
change on the distribution of indigenous forest in KwaZulu-Natal, South
Africa. J. Biogeogr. 26: 595–617.
Felton, I.R. 2002. A strategic assessment of the woody grasslands of
Maputaland, South Africa, with recommendations for future conservation.
M.Sc. thesis, Institute of Geography and Earth Sciences, Univ. of Wales,
Aberystwyth.
Glen, H.F. 1996. A description of the vegetation of Oribi Gorge Nature
Reserve, Natal. Report, National Botanical Institute, Pretoria.
Goldberg, K. & Frank, A. 2004. Maputaland-Pondoland Bushland and
Thickets. In: Burgess, N., D’Amico Hales, J., Underwood, E., Dinerstein, E.,
Olson, D., Itoua, I., Schipper, J., Ricketss, T. & Newman, K. (eds), Terrestrial
ecoregions of Africa and Madagascar—a conservation assessment, pp.
360–362. Island Press, Washington.
Heijnis, C. 2004. KwaZulu–Cape Coastal Forest Mosaic. In: Burgess, N.,
D’Amico Hales, J., Underwood, E., Dinerstein, E., Olson, D., Itoua, I.,
Schipper, J., Ricketss, T. & Newman, K. (eds), Terrestrial ecoregions of
Africa and Madagascar—a conservation assessment, pp. 258, 259. Island
Press, Washington.
Hobday, D.K. & Mathew, D. 1974. Depositional environment of the Cape
Supergroup in the Transkei. Trans. Geol. Soc. S. Afr. 77: 223–227.
Hoffman, M.T. 1983. The grasslands of the Hluleka Nature Reserve. Report,
Dept of Botany, Univ. of Cape Town.
Hueck, K. 1966. Die Wälder Südamerikas. G. Fischer, Stuttgart.
Huntley, B.J. 1984. Characteristics of South African biomes. In: De Booysen,
P.V. & Tainton, N.M. (eds), Ecological effects of fire in South African ecosys-
tems, pp. 1–17. Springer, Berlin.
Kirkwood, D. 2004. Maputaland Coastal Forest Mosaic. In: Burgess, N.,
D’Amico Hales, J., Underwood, E., Dinerstein, E., Olson, D., Itoua, I.,
Schipper, J., Ricketss, T. & Newman, K. (eds), Terrestrial ecoregions of
Africa and Madagascar—a conservation assessment, pp. 256, 257. Island
Press, Washington.
Lawes, M.J. 1990. The distribution of the samango monkey (Cercopithecus
mitis erythrarchus Peters, 1852 and Cercopithecus mitis labiatus I. Geoffroy,

582
STRELITZIA 19 (2006)
Indian Ocean Coastal Belt
1843) and forest history in southern Africa. J. Biogeogr. 17: 669–680.
Le Roux, N.P. 1995. Grassland of Umtamvuna Nature Reserve, KwaZulu-
Natal: a description and recommendation for monitoring. M.Sc. thesis,
Univ. of Natal, Pietermaritzburg.
Low, A.B. & Rebelo, A.(T.)G. (eds) 1996. Vegetation of South Africa,
Lesotho and Swaziland. A companion to the vegetation map of South
Africa, Lesotho and Swaziland. Dept of Environmental Affairs and Tourism,
Pretoria.
Lubbe, R.A. 1997. Vegetation and flora of the Kosi Bay Coastal Forest
Reserve in Maputaland, northern KwaZulu-Natal, South Africa. M.Sc. thesis,
Dept. of Botany, Univ. of Pretoria.
Marloth, R. 1907. On some aspects in the vegetation of South Africa which
are due to prevailing winds. Rep. S. Afr. Assoc. Advanc. Sci. 1905/1906:
215–218.
Matthews, W.S., Van Wyk, A.E. & Van Rooyen, N. 1999. Vegetation of the
Sileza Nature Reserve and neighbouring areas, South Africa, and its impor-
tance in conserving the woody grasslands of the Maputaland Centre of
Endemism. Bothalia 29: 151–167.
Maud, R.R. & Botha, G. 2000. Deposits of the South Eastern and Southern
Coasts. In: Partridge, T.C. & Maud, R.R. (eds), Cenozoic of southern Africa,
pp. 19–32. Oxford Univ. Press, Oxford.
Mazus, H. 2000. Clues on the history of Podocarpus forest in Maputaland,
South Africa, during the Quaternary, based on pollen analysis. Afr. Geosci.
Rev. 7: 75–82.
Midgley, J.J. 1986. Are 800 km disjunction in Leucadendron due to earlier
climatic, tectonic or edaphic changes or long distance dispersal? S. Afr. Soc.
Quater. Res. 1: 51, 52.
Moll, E.J. 1972. The distribution, abundance and utilization of the Lala palm,
Hyphaene natalensis, in Tongaland, Natal. Bothalia 10: 627–636.
Moll, E.J. 1978. The vegetation of Maputaland. Trees in South Africa 29:
31–58.
Moll, E.J. 1980. Terrestrial plant ecology. In: Bruton, M.N. & Cooper K.H.
(eds), Studies on the ecology of Maputaland, pp. 52–68. Rhodes University,
Grahamstown & Natal Branch of Wildlife Society of Southern Africa,
Durban.
Moll, E.J. & White, F. 1978. The Indian Ocean Coastal Belt. In: Werger, M.J.A.
& Van Bruggen, A.C. (eds), Biogeography and ecology of southern Africa,
pp. 561–598. Dr W. Junk, The Hague.
Müller, M.J. 1982. Selected climatic data for a global set of standard stations
for vegetation science. Dr W. Junk, The Hague.
Murray, B.G., Ran, Y., De Lange, P.J., Hammett, K.R.W., Truter, J.T. &
Swanevelder, Z.H. 2004. A new species of Clivia (Amaryllidaceae) endemic
to the Pondoland Centre of Endemism, South Africa. Bot. J. Linn. Soc. 146:
369–374.
Myre, M. 1964. A vegetacão do extremno sul de provincia da Moçambique.
Junta de Investigaçoes do Ultramar, Lisboa.
Myre, M. 1971. As pastagens da regiao da Maputo. Mémorias Instituto de
Investigaçao Agronómica de Moçambique 3: 1–181.
Oschaldeus, H.D., Vogel, J.C. & Scott, L. 1996. Radiometric date for the Port
Dunford peat and development of yellow-wood forest along the South
African east coast. S. Afr. J. Sci. 92: 43–45.
Partridge, T.C., Avery, D.M., Botha, G.A., Brink, J.S., Deacon, J., Herbert,
R.S., Maud, R.R., Scholtz, A., Scott, L., Talma, A.S. & Vogel, J. 1990. Late
Pleistocene and Holocene climatic change in southern Africa. S. Afr. J. Sci.
86: 302–306.
Por, F.D. 1992. Sooretama: the Atlantic forest of Brazil. SPB Publ., The
Hague.
Prell, W.L., Hutson, W.H., Williams, D.F., Bé, A.W.H., Geitzenaur, K. &
Molfino, B. 1980. Surface circulation of the Indian Ocean during the last
glacial maximum, approximately 18,000 yr. B.P. Quater. Res. 14: 309–336.
Ramsey, P.J. 1994. Marine geology of the Sodwana Bay shelf, South Africa.
Mar. Geol. 120: 225–247.
Riegl, B. 2003. Climate change and coral reefs: different effects in two high-
altitude areas (Arabian Gulf, South Africa). Coral Reefs 22: 433–446.
Rutherford, M.C. & Westfall, R.H. 1986. Biomes of southern Africa—an
objective categorization. Mem. Bot. Surv. S. Afr. No. 54: 1–98.
Roberts, D.C. 1993. The vegetation ecology of municipal Durban, Natal.
Floristic classification. Bothalia 23: 271–326.
Schipper, J. & Burgess, N. 2004a. Northern Zanzibar-Inhambane Coastal
Forest Mosaic. In: Burgess, N., D’Amico Hales, J., Underwood, E.,
Dinerstein, E., Olson, D., Itoua, I., Schipper, J., Ricketss, T. & Newman, K.
(eds), Terrestrial ecoregions of Africa and Madagascar—a conservation
assessment, pp. 252–254. Island Press, Washington.
Schipper, J. & Burgess, N. 2004b. Southern Zanzibar-Inhambane Coastal
Forest Mosaic. In: Burgess, N., D’Amico Hales, J., Underwood, E.,
Dinerstein, E., Olson, D., Itoua, I., Schipper, J., Ricketss, T. & Newman, K.
(eds), Terrestrial ecoregions of Africa and Madagascar—a conservation
assessment, pp. 254–256. Island Press, Washington.
Scott, L., Coremans, B. & Maud, R.R. 1992. Preliminary palynological evalu-
ation of the Port Dunford Formation, Natal, South Africa. S. Afr. J. Sci. 88:
470–474.
Scott, L. & Steenkamp, M. 1996. Environmental history and recent human
influence at coastal Lake Teza, KwaZulu-Natal. S. Afr. J. Sci. 92: 348–350.
Scott-Shaw, R. 1999. Rare and threatened plants of KwaZulu-Natal and
neighbouring regions. KwaZulu-Natal Nature Conservation Service,
Pietermaritzburg.
Shackleton, C.M. 1989. The grassland dynamics of Mkambati Game Reserve.
M.Sc. thesis, Univ. of Witwatersrand, Johannesburg.
Shackleton, C.M. 1992. Area and species selection by wild ungulates in coastal
sour grasslands of Mkambati Game Reserve. Afr. J. Ecol. 30: 189–202.
Shackleton, C.M., Granger, J.E., McKenzie, B. & Mentis, M.T. 1991.
Multivariate analysis of coastal grasslands at Mkambati Game Reserve,
north-eastern Pondoland, Transkei. Bothalia 21: 91–107.
Shackleton, S.E. 1990. Socio-economic importance of Cymbopogon validus
in Mkambati Game Reserve. S. Afr. J. Bot. 56: 675–682.
Shackleton, S.E. & Shackleton, C.M. 1994. Habitat factors influencing the
distribution of Cymbopogon validus in Mkambati Game Reserve, Transkei.
Afr. J. Range For. Sci. 11: 1–6.
Smith, R.J. 2001. Designing an integrated protected area network for
Maputaland. Ph.D. thesis, Durell Institute of Conservation Ecology, Univ.
of Kent, Canterbury, UK.
Stokes, S., Thomas, D.S.G. & Washington, R.W. 1997. Multiple episodes
of aridity in southern Africa since the last interglacial Period. Nature 388:
154–159.
Thomas, R.J., Marshall, C.G.A., Du Plessis, A., Fitch, F.J., Miller, J.A., Von
Brunn, V. & Watkeys, M.K. 1992. Geological studies in southern Natal and
Transkei: implications for the Cape Orogeny. In: De Wit, M.J. & Ransome,
I.G.D. (eds), Inversion tectonics of the Cape Fold Belt, Karoo and Cretaceous
basins of southern Africa, pp. 229–236. A.A. Balkema, Rotterdam.
Tinley, K.L. 1985. Coastal dunes of South Africa. S. Afr. Natl. Sci. Progr. Rep.
No. 109: 1–300.
Tyson, P.D. 1986. Climatic change and variability in southern Africa. Oxford
Univ. Press, Cape Town.
Van der Linden, J.P., Siebert, S.J., Siebert, F., Ferreira, D.P. & Bredenkamp, G.J.
2005. Vegetation of the Owen Sithole College of Agriculture in Zululand,
KwaZulu-Natal. Koedoe 48: 33–56.
Van Wyk, A.E. 1989. Floristics of the Natal/Pondoland sandstone forests. In:
Geldenhuys, C.J. (ed.), Biogeography of the mixed evergreen forests of
southern Africa, pp. 145–157. FRD, Pretoria.
Van Wyk, A.E. 1990a. The sandstone regions of Natal and Pondoland:
remarkable centres of endemism. Palaeoecology of Africa 21: 243–257.
Van Wyk, A.E. 1990b. A new species of Leucadendron (Proteaceae) from
Pondoland, with a discussion of its biogeography. S. Afr. J. Bot. 56: 458–466.
Van Wyk, A.E. 1994. Indian Ocean Coastal Belt: CPD Site AF59. Maputaland-
Pondoland Region, South Africa, Swaziland and Mozambique. In: Davis,
S.D., Heywood, V.H. & Hamilton, A.C. (eds), Centres of plant diversity. A
guide and strategy for their conservation. Volume 1: Europe, Africa, South
West Asia and the Middle East, pp. 227–235. IUCN Publ., Cambridge.
Van Wyk, A.E. 1996. Biodiversity of the Maputaland Centre. In: Van der
Maesen, L.J.G., Van der Burgt, X.M. & Van Medenbach de Rooy, J.M. (eds),
The biodiversity of African plants, pp. 198–207. Kluwer, Dordrecht.
Van Wyk, A.E. & Smith, G.F. 2001. Regions of floristic endemism in southern
Africa: a review with emphasis on succulents. Umdaus Press, Pretoria.
Van Zinderen Bakker, E.M. 1982. African palaeoenvironments 18,000 yrs BP.
Palaeoecology of Africa 15: 77–99.
Venter, H.J.T. 1972. Die plantekologie van Richardsbaai, Natal. D.Sc. thesis,
Dept of Botany, Univ. of Pretoria.
Vlok, J.H.J. & Euston-Brown, D.I.W. 2002. Subtropical Thicket Ecosystem
Planning Project (STEP). Biological Survey Report (plants and birds). Report,
Terrestrial Ecology Research Unit, Univ. of Port Elizabeth. www.zoo.upe.
ac.za/step
Vlok, J.H.J., Euston-Brown, D.I.W. & Cowling, R.M. 2003. Acocks’ Valley
Bushveld 50 years on: new perspectives on the delimitation, characteriza-
tion and origin of subtropical thicket vegetation. S. Afr. J. Bot. 69: 27–51.
Von Maltitz, G., Mucina, L., Geldenhuys, C.J., Lawes, M.J., Eeley, H., Adie,
H., Vink, D., Fleming, G. & Bailey, C. 2003. Classification system for South
African indigenous forests. An objective classification for the Department
of Water Affairs and Forestry. Report ENV-PC 2003-17, Environmentek,
CSIR, Pretoria.
Ward, C.J. 1980. The plant ecology of the Isipingo beach area, Natal, South
Africa. Mem. Bot. Surv. S. Afr. No. 45: 1–147.
Weisser, P.J. 1978. Conservation priorities in the dune area between Richards
Bay and Mfolozi Mouth based on vegetation survey. Natal Town Region.
Plan. Rep. No. 38: 1–64.
Weisser, P.J. 1987. Dune vegetation between Richards Bay and Mlalazi
Lagoon and its conservation priorities in relation to dune mining. Natal
Town Reg. Plan. Suppl. Rep. No. 19: 1–71.
Weisser, P.J. & Marques, F. 1979. Gross vegetation changes in the dune area
between Richards Bay and the Mfolozi River. Bothalia 12: 711–721.
Wessels, N.G. 1991. The syntaxonomy and synecology of swamp forests in

STRELITZIA 19 (2006)
583
Indian Ocean Coastal Belt
the Lake St. Lucia area, Natal. Report, Ecosystem Management Programme,
Division of Forest Science and Technology, CSIR, Pretoria.
White, F. 1976. The underground forests of Africa: a preliminary review. The
Gardens Bulletin (Singapore) 29: 55–71.
White, F. 1983. Vegetation of Africa: a descriptive memoir to accompany the
UNESCO/AETFAT/UNSO vegetation map of Africa. UNESCO, Paris.
Wright, C.I., Miller, W.R. & Cooper, J.A.G. 2000. The late Cenozoic evolu-
tion of coastal water bodies in northern KwaZulu-Natal, South Africa. Mar.
Geol. 167: 207–229.
