Mapping Salt-affected Soils Using Remote Sensing Indicators - A Simple Approach With the Use of GIS IDRISI -

摘要

This research deals with the problem of monitoring saline soils of Faisalabad, Pakistan. The analysis is based on remote sensing data of LISS-II sensor of IRS-1B satellite using GIS-IDRISI functions. We have examined how different remote sensing indicators work for salt-affected soil classification in the study area. The study has suggested some new but simple and practicle approaches for the problem. We have analyzed the effectiveness of several indicators for the presence of salts in the area: salinity indices, normalized differential salinity index, and ratio of the signals received by the sensor in the 3rd spectral band to others. As salt-affected soils are characterized also by stressed vegetation, we therefore have analyzed vegetation indices concurrently. The probability of correct classification of the satellite image was observed strongly dependent on the season for all indicators analyzed. The best results can be achieved for the dry season (March-April) but not in humid or high temperate periods which may create confusion some times with other classes, specially settlements areas. The most difficult part in the classification processes was to distinguish between soil-affected areas and populated areas, which has muddy roofs similar to dry barren soils along with patchy saline areas within the village. We have come-up with two original schemes of classification through the analysis of the available data for this specific area. The first one uses COMPOSITE and STRETCH modules to produce two new images and then analyze their ratio. The COMPOSITE module takes data received in 1st, 2nd and 4th channels and STRETCH function produces histogram equalization contrast stretch of the 4th channel data. The second scheme uses ISOCLUST function of GIS IDRISI that performs classification based on specifically created images (through salinity indices and PCA analysis) instead of common practice of using just satellite measurements. Both schemes are shown to be able to perform good classification of the study area.