MADAGASCAR:
Vegetation Mapping and Biodiversity Conservation
(using Geographical Information Systems)
Madagascar
is singled out by the international scientific
and conservation community as one of the richest
countries in the world in terms of biodiversity,
endemism and range of habitats. Its flora is
diverse and unique. Of approximately 10,000
native higher plant species, about 8,000 species
are thought to be endemic to the island. As a
comparison, Madagascar is about 2.5 times as
large as Britain, which has about 1,200 species
of which only 10 to 20 are endemic. The value of
the flora of Madagascar, both to the local
peoples and in a global sense, is potentially
immense. Despite its importance, this flora is
under serious threat. Over 80% of the island has
already been stripped of its native vegetation
cover; the majority of this area is now very
species-poor secondary grassland which is burnt
annually and is subject to intense erosion. The
heritage of biological diversity in Madagascar
is probably under greater threat than in any
other country. This unique diversity, combined
with the threats to the remaining native
vegetation, puts Madagascar amongst the highest
conservation priority areas in the world.
In
response to the Convention on Biological
Diversity (resulting from the ‘Rio Summit’),
a conservation strategy is being implemented as
part of Madagascar's Environmental Action Plan:
part of this Action Plan is to increase the
number of protected areas. This project has
aimed at providing assistance to the
organisations involved in this new conservation
impetus, through the prioritisation of key
regions to include in the new network of
reserves, and to provide a powerful conservation
planning tool with all maps and data available
and being used in Madagascar.
Our
GIS studies in Madagascar are primarily aimed at
analysing distribution patterns of vegetation
types and plant biodiversity in Madagascar.
Field work has shown that the structure and
species composition of the vegetation often
alters radically with changes in substrate. It
is assumed, therefore, that different vegetation
types, with different species compositions,
occur on different rock types, and that a new,
more informative map of vegetation types could
be produced by subdividing the broad primary
vegetation types on the basis of the rock type
on which they occur.
The
first stage was to digitise the geology map of
Madagascar (BESAIRIE, 1964), and then to
simplify it into broad categories of rock types
which seem to have an important effect on the
vegetation they support. The categories of
sedimentary rocks include sandstones, loose
(unconsolidated) sands, and two limestone
categories of different ages (one of which
produces the spectacular "tsingy"
areas of jagged, highly eroded limestone
pinnacles). A broad category of metamorphic
rocks (including granites and migmatites), often
covered by thick layers of laterites, covers
large areas of the central and eastern areas of
the island. Lavas and basalts, and several
restricted rock types including quartzites,
marbles and ultrabasics are also separated.
The
remaining areas of good quality, primary
vegetation were identified from the vegetation
map of FARAMALALA (1988 &1995), which was
derived from LANDSAT satellite imagery, and
broadly maintained the vegetation zones defined
by HUMBERT (1955). The primary vegetation was
then classed into two major types, one evergreen
(eastern and central) and the other deciduous
(western and southern). The evergreen forests
include the humid forests (at low, medium and
high altitudes), along with the sclerophyllous
woodland remnants (Uapaca dominated) of central
Madagascar. The deciduous formations include the
seasonally dry forests of the west and north,
and the southern, dry forests and scrublands.
A map of the ‘Remaining
Primary Vegetation, classified by the Underlying
Geology’ was produced by overlaying the
map of ‘Remaining
Primary Vegetation’ (derived from
satellite imagery) on a map of ‘Simplified
Geology’. This map provides new insights
into the patterns of distributions of plant
species, and the various types of vegetation,
especially in western
and southern
Madagascar where the geology varies
substantially.
Click
on Maps for More Information and Larger Images
Histograms
of the area remaining of each vegetation type
(in km2) could then be produced, and by
overlaying a map of the protected
areas (COEFOR/CI, 1993), the amount of
protection for each type could be demonstrated.
It is immediately obvious which vegetation types
are poorly represented in the current system of
protected areas. The maps can then be
re-examined to show where large, intact areas
suitable for conservation still exist. If
Reserves were set up on each category, then the
system of reserves would include as wide a range
of vegetation types as possible, and therefore
as high a species diversity as possible. When
used in this way, as a mirror of plant
diversity, this is obviously an important tool
for conservation planning and management.
The
graphs and maps output have been used to
succinctly convey complex arguments in an
attractive, user-friendly format accessible to
non-botanists, allowing politicians and decision
makers to utilise the data. It also provides
statistically based arguments for conservation
priorities in an area which has previously been
dominated by intuition, or based on the
distributions of a few, well-known species. It
is clearly evident, for example, that the
deciduous, dry, southern forests and scrublands
are insufficiently protected, especially given
their recognised richness in endemic plant
species, and the great variability in the
underlying geology.
These
maps and data have been compiled and distributed
on CD-ROM, to the different organisations
involved in conservation projects in Madagascar,
including ANGAP (the National Association for
the Management of Protected Areas), WWF
and Conservation
International. They are now being used in
Madagascar, within the context of the
Environmental Action Plan (a response to
Madagascar signing the Convention on
Biodiversity in March, 1996), to help identify
areas of high priority for conservation of
biodiversity, and to improve the network of
Protected Areas.
Following
the success of this initial work, it is
envisaged that similar approaches will be used
in other regions, and that the methodology could
be standardised, perhaps using other sources of
digitised map data (such as meteorological
satellite data for climate or seasonality).
Back
to top
References:
- BESAIRIE,
H. (1964). Carte Géologique de Madagascar,
au 1:1,000,000e, trois feuilles en couleur.
Service Géologique, Antananarivo.
- COEFOR/CI
(1993). Répertoire et Carte de Distribution
: Domaine Forestier de Madagascar. Direction
des Eaux et Forêts, Service des Ressources
Forestières, Projet COEFOR (Contribution à
l’étude des Forêts Classées), et
Conservation International,
20 pp. + 1 map
- DU
PUY, D.J. and MOAT, J.F. Vegetation
Mapping and Classification in Madagascar
(using GIS): Implications and
Recommendations for the Conservation of
Biodiversity. In: Cutler, D.F., Huxley,
C.R. & Lock, J.M. (Eds.), The ecology,
chorology and taxonomy of the African and
Madagascan floras - Proceedings of the
Frank White Memorial Symposium. Kew Bulletin
Additional Series. Royal Botanic Gardens,
Kew.
- DU
PUY, D.J. and MOAT, J. (1996). A
refined classification of the primary
vegetation of Madagascar based on the
underlying geology: using GIS to map its
distribution and to assess its conservation
status. In: W.R. Lourenço (editor).
Proceedings of the International Symposium
on the Biogeography of Madagascar, pp.
205--218, + 3 maps. Editions de l’ORSTOM,
Paris.
- FARAMALALA,
M.H. (1988). Etude de la Végétation de
Madagascar à l’aide des Données
spatiales. Doctoral Thesis, Univ. Paul
Sabatier de Toulouse, 167 p. + map at
1:1,000,000.
- FARAMALALA,
M.H. (1995). Formations Végétales et
Domaine Forestier National de Madagascar.
Conservation International (et al.), 1 map.
- HUMBERT,
H. (1955). Les Territoires Phytogéographiques
de Madagascar. Leur Cartographie. Colloque
sur les Régions Ecologiques du Globe, Paris
1954. Ann. Biol. 31: 195-204, + map.
Back
to top
Some
Useful Links
Back
to top
|
|