Modeling for Global Land Degradation
using Remote Sensing and GIS
Krishna Pahari, Shunji Murai
Space Technology Applications and Research program
School of Environment, Resources and Development,
Asian Institute Of Technology, GPO Box 2754,
Bangkok 10501, Thailand
Abstract
Land degradation or desertification has become one of most pressing environmental problems in the world today. This paper is an attempt to highlight the use of GIS and remote sensing for the assessment of global land degradation. The research is based on a combination of climatic modelling, mainly based on rainfall and temperature data and the physical process modelling, based on the analysis of land degradation mechanics, for which land use is an important contributing factor. Some preliminary findings of the research are also presented.
1. General Background
Land degradation is the reduction of land resource potential which can be manifested in various forms such as soil degradation (waer erosion, wind erosion or chemical degradation), degradation of vegetation of degradation of rangeland. Desertification may be considered as the land degradation in the arid, semiarid and dry subhumid areas due to natural of anthropogenic factors.
Although the land degradation is one of the most pressing environmental problems at the global level, the systematic study of global land degradation or desertification is emerging only recently. The technology of Remote Sensing together with global GIS is providing new possibility for a more scientific analysis of this problem.
2. The Research Methodology
In this research, the extent and nature of desertification is being studied by using two approaches, namely climate modeling based mainly on rainfall and temperature data, and the physical or bio-process modeling in which remote sensing data an important input.
2.1 Climatic model
This mainly consists of modeling for aridity and moisture indices calculated on the basis of temerature and rainfall data from many points around the globe and then interoplated to the areal data. The various indices considered are the following:
i.
Moisture Index
This index is calculated by :
Moisture Index = Annual Rainfall/PET,
where, PET is the potential evapotranspiration, which can be estimated by using either thormthwaite method, Holdridge method or others. So for, the PET data produced by Ahn and Tateishi has been used for calculating this index.
ii.
Martonne's Aridit Index
This is given by:
where, P = annual precipitation (mm),
T = sum of monthly mean temperature of those months with monthly mean temperature greater then 0, divided by 12.
on the basis of the values of this index, different areas can be classified into various zones.
iii.
Aridity Index revised by Murai and Honda
The index used by Murai and Honda (1991) is slightly revised from the Martonne's index, with additional distinction between ordinary forest and tropical forest.
The range of values for different aridity classes in this method is given in Table 1. Figure 1 illustrates the steps involved in modeling aridity zones.
Table 1 Aridity indices for different classes
| Class |
Martonne AI |
Revised AI (Murai and Honda) |
| Desert |
<=5 |
<=5 |
| Semi-desert |
5-10 |
5-10 |
| Grass land |
10-30 |
10-20 |
| Forest |
>30 |
20-40 |
| Tropical forest |
-- |
>40 |
| (annual ave temp>24°C) |
Figure 1. Flow chart of methodology for climatic modeling of aridity zones.
