Estimation of Soil Salinity Components Using Radiance andReflectance Transformations of ASTER and ETM+ Imagery(Case Study: Abarkooh, Yazd)

Salinity monitoring is one of the most important environmental issues especially in arid and semi-arid regions. To this end, it is inevitable to use modern instruments such as remote sensing. The main objective of the present study is to estimate soil salinity parameters using spectral radiance and reflectance transformations of ETM+ and ASTER images in 87.4 Km2. of Abarkooh Desert (located in Yazd Province). To do so, ETM+ (October 2002), ASTER (June 2006) images and chemical analysis of 29 filed soil samples (collected in June 2006) were used. At first, the salinity components of soil (EC, CL, and, SO4) were estimated in a stepwise regression modeling through applying of spectral transformation and reflectance on satellite images of ETM+ and ASTER. The estimationmodels were evaluated using R2, RE, RMSE and CE and validated using the results derived from 9 sample points of the field samples as independent test objects. The results indicate that applying the transformations has different effects on increasing the model evaluation criteria for different components. Generally, by applying this type of transformations on the images, the accuracy of estimation soil salinity parameters (for instance, correlation coefficient has increased 35% in estimation of CL using ETM+ and 10% in estimation of EC using ASTER image) has increased.