Assessment of Satellite Images for Soil Salinity Studies

Soil salinity is one of the main environmental problems affecting extensive area in the world. There are some problems with traditional data collection methods for soil studies. Using the new methods and techniques such as remote sensing could overcome most of these problems. However using these data in areas with uncommon usages needed some researches to find the best calibration between the data and real situations in soil. This research was carried out using Landsat satellite images in Neyshabour area, North East of Iran. In order to prepare suitable learning samples for the image processing in this study, 300 locations were randomly selected in the area, among which 273 locations were finally selected as suitable surface soil samples. All samples were moved to laboratory and their electrical conductivity was measured. Six reflective bands of ETM+ satellite images taken from the study area in 2002 were used for the image processing analysis. Classification of different soil salinities was carried out using common algorithms of image classification based on the best composition bands and using statistical methods between soil salinity variables and digital numbers of the images to represent a suitable method. the research results showed that the reflective bands 7, 3, 4 and 1 are the best band composition for preparing the color composite images and for the classification of the salinity in this area. The highest coefficient of determination was R2=0.311 and R2=0.44 for saline and non-saline soil respectively using band 2 and 3 of the images at 5% significant level. Based on the results, it can be concluded that the potential of ETM+ images for delineation and identification of different soil salinity are limited.