Thesis (M. Sc.) - Asian Institute of Technology, 1996

A thesis submitted in partial fulfilment of the requirements for the degree of Master of Science, School of Environment, Resources and Development

Salinity is highly detrim ental to both ferti lity of the so il and the environment. Assessment of soil salinity and its various physical factors in a timely and cost effective manner to reclaim the salt-affected area is one of the prime requirements in the progress and development of the agri culturally dependent countries like Pakistan. Remote Sensing (RS) and Geographic Information System (GIS) technologies are emerging as indispensable tools in the study and mapping of this dynamic phenomenon. This study presents a workable methodology with resu lts, obtained through application of RS data in integration of GIS, to assess soil salinity and its ph ys ical constraints. Approximately 788 sq. km. of salt-affected area in an irrigated regime of southern Punjab, Pakistan was selected for the stud y. Digi tal Image Processing and Classification and stati s1ical techniques were used in estab li shing objective-ori en ted results. An extensive ground truth survey of the study area was conducted to collect first hand knowledge about the problem extent of salinity in the area The usefulness of Landsat Thematic Mapper (TM) and SPOT Multispectral (XS) data has been assessed through qu ali tative and quantitative anal ys is. The capability and limi tations of sensors were studi ed on the basis of resultant statistics of the training samples used for classification . A highly saline area, mostly in crust form , is eas ily identifiable and clearly distingui shable from other classes. However, an accurate differentiation between moderately salin E' and other highly refl ective fine eroded materi al is not possible with existing resolution of the remote sensing data and capabi lities of available image processing systems. The vegetation having high density and canopy is also an hindrance in discrimination of slightly salinE' from non sal ine areas. Based on the res ults of the study, it can be concluded that integrated use of RS and GIS i:; highly useful in assessment of soil sa linity and some of its related factors.