8 
ECOLOGICAL ANALYSIS AND MODELLING : ISSUES AND CHALLENGES 
Harini Nagendra' and Madhav Gadgil 1 ’ 2 
Centre for Ecological Sciences, Indian Institute of Science, Bangalore - 560012 
2 .la\vaharlal Nehru Centre for Advanced Scientific Research. Jakkur P.O., Bangalore - 560064 
ABSTRACT 
India is a country with rich biological resources, which are being depleted over time. Field measurements will not be sufficient to 
monitor changes in these resources over time, and must be complemented by other means of assessment, namely remote sensors. 
Linkages need to be established between extensive remotely sensed information derived at a broader scale and intensive, locally 
collected field data. Ecosystem mapping provides a means for this. Landscapes can be classified into different ecosystem types 
using remotely sensed data, and each type associated with specific environmental parameters (such as a certain set of species or 
levels of soil erosion) based on field studies. Preliminary investigations in the Western Ghats and west coast of India have 
demonstrated the feasibility of this methodology for assessing and monitoring Angiosperm species diversity. A similar 
methodology can be extended to monitoring other natural resources such as biomass and land and water quality. However, crucial 
details of the methodology to be defined are the scale at which this classification is to be carried out, the number of ecosystem 
types to be differentiated, the nature and extent of field information, and the extent of correlation between broad scale 
classification and detailed field information, for different situations. Taking all these aspects into consideration, and based on our 
earlier experience in using such a methodology for biodiversity assessment, a multi-scale programme of biological resource 
assessment and monitoring across the Western Ghats is proposed. 
BACKGROUND 
The structure, composition and distribution of 
natural resources on earth has never been static - there 
have always been changes in land and water quality, 
species composition, landscape structure and other 
natural resources. This is part of the nature of life itself — 
but human activities of various kinds have severely 
accelerated the rates of changes in these environmental 
parameters. This poses problems of several kinds, 
including purely economic difficulties, health hazards, 
and even less obvious issues like loss in functionality - 
for example, studies suggest that when biodiversity 
levels fail beiow certain minimum requirements, entire 
ecosystems may collapse. These and other factors make 
the slowing of rates of depletion of the earth’s natural 
resources critical. 
For this, we need to identify major forces driving 
these losses, and determine where these processes are 
most active and losses maximal. This information can be 
used to drive strategies for resource conservation. 
Several such studies have been carried out, but 
satisfactory answers still elude us. This is not for lack of 
effort, nor is it due to a lack of scientific capability. The 
problems associated with the distribution and decline of 
natural resources are extremely multifaceted and do not 
permit neat and simple studies. 
Even specifically for India, this task poses several 
important scientific and logistic challenges. Ours is a 
country of rich and varied ecosystems and habitats, 
harboring a virtual treasure house of natural resources. 
These are distributed over a vast 320 million ha of land 
and 200 million ha of exclusive economic zone in the 
sea. To inventory, monitor and conserve natural 
resources over such a large and varied area is an 
enormous task. 
REMOTE SENSING BASED NATURAL 
RESOURCE ASSESSMENT 
Remote sensing has a high potential capability for 
natural resource assessment, as it can provide 
information about the distribution and composition of 
several kinds of natural resources like water, soil, 
minerals, biomass and vegetation cover, over large areas 
at a glance. Ecologists in fact have investigated the 
potential of remote sensing for various studies, mainly 
for land use and land cover mapping (Roughgarden et 
ah, 1991; Turner et al., 1994). However, given the size 
of the country and the fact that the scale of variation in 
distribution of resources is often too small to be picked 
up by remote sensors at commercially available 
resolutions, it is not possible to completely census 
resources using only remotely sensed data. 
Supplementary data in the form of field observations are 
required. 
Data collected by remote sensors is extensive, and 
at relatively coarse resolutions. Field data, for example 
on water quality, soil erosion or species distributions, 
cannot be as extensive and must be sampled at various