12 
and organizations for the Western Ghats and west coast 
of India, and the information that these maps and tables 
contain a specific parameter namely species diversity. 
The French Institute of Pondicherry in collaboration 
with the Indian Council for Agricultural Research has 
published a vegetation map of Peninsular India at a 
1:1,000,000 scale. Two sheets of this map, Cape 
Comorin and Mysore (Gaussen et al., 1961, 1965) cover 
the entire Western Ghats area. Subsequently, the Forest 
Map of South India has been brought out, at a 1:250,000 
scale (Pascal, 1982, 1986). Maps of the Nilgiri 
Biosphere Reserve area have also been made, to a 
detailed 1:100,000 scale (Prabhakar and Pascal, 1994). 
The classification schemes characterize ecosystems 
based on their plant cover, and identify them with the 
help of visual interpretation of Landsat MSS and TM, 
and SPOT imagery. Fifteen types have been recognized 
in the evergreen and semi-evergreen climax forests, and 
three types in the deciduous climax forests. Based on a 
few field plots, for each of the natural types as well as 
some of their degradation stages, detailed floristic lists 
are provided (Pascal, 1986). These lists are not 
obviously exhaustive - however, for some of the more 
natural types where more plots have been laid, they 
estimate that they have encountered between 75% and 
90% of the species present in that type. However, as no 
analysis has been published demonstrating that these 
types differ significantly in their species composition, it 
is difficult to extrapolate this information to other 
mapped areas where no field investigations have been 
made. 
The Forest Survey of India in collaboration with 
the National Remote Sensing Agency of India has 
carried out a 1:1,000,000 scale mapping of forest cover 
in India several times, the most recent being during 
1989-91 (Dutt et al., 1994). Visual interpretation of 
Landsat MSS imagery has been used to prepare these 
maps. The categories employed are however very few - 
closed forest, open/degraded forest, mangrove forest and 
non-forest (which encompasses agricultural areas, 
barren land and non-forest plantations). Forest 
plantations such as Acacia auriculiformis are classified 
under the forest categories. 
This kind of mapping is obviously not very useful 
for assessing species composition or diversity. Flowever, 
in 1993-1994, a detailed, 1:25,000 scale classification of 
the Western Ghats area into different types of forests - 
evergreen, semi-evergreen and deciduous - with varying 
levels of canopy cover discriminated - as well as 
plantations, crops and other ecosystem types has been 
carried out. This map has not been so far published. 
The National Remote Sensing Agency has 
prepared wasteland maps for several parts of India, 
especially those dominated by barren lands. This is 
perhaps the only large scale mapping effort in the 
Western Ghat region mapping ecosystems that lack 
vegetation cover. These maps are at a 1:50,000 scale and 
have been prepared using visual interpretation of 
Landsat MSS imagery. Thirteen categories are defined 
(NRSA, 1991). As is apparent from a look at these 
categories, this system of classification combines land 
use systems like grazing or shifting cultivation and 
physical features such as ravinous lands or upland 
characteristics along with ecosystem types such as 
agricultural or plantation areas. This set of maps does 
not provide information about species diversity. 
However, since this classification scheme is created for 
the purpose of wasteland reclamation, it would have 
been useful if it also provided information on natural 
resource parameters such as soil erosion rates, or water 
quality in each type. This would permit better 
understanding of the different wasteland types, for their 
management. 
Recently, a more detailed system of classification 
for India has been proposed by the Space Applications 
Centre under the National (Natural) Resources 
Information System (NRIS) (SAC, 1997). This scheme 
is recommended for use while mapping India's natural 
resources - forests, agricultural areas, wastelands and 
cities, and the categories used integrate information 
obtained from remote sensing, conventional 
cartographic maps as well as non-spatial data on 
parameters like demography. The categories are not 
suited to natural resource distribution analysis, and 
surely not for species diversity analysis. All forest 
plantations, for example, are clubbed under one 
category. Evergreen as well as semi-evergreen forests 
are also combined. They define ‘open’ and ‘closed’ 
forests, and it is not clear what the delineating line is 
between these two, or between open and scrub forests. 
Crops, also, are classified as kharif and rabi, depending 
on the growing season, rather than based on the species 
or set of species grown. 
These classification schemes in their present form 
do not provide complete data for the management of 
natural resources. If these extensive, remote sensing 
based maps can be standardized and correlated with the 
distribution of natural resources on the ground, then this 
will provide a vast store of information which can be 
used to plan effective management strategies for 
harvesting natural resources. 
One such attempt has been carried out by Ramesh 
et al. (1997). This provides an elaborate analysis of the