Mapping Salinity and Sodicity of Surface Soil by Remote Sensing and Geostatistic Methods in South Side of Mah Valat County

آب و خاک

Article 7, Volume 25, Issue 3, July 2011   

Document Type: Research Article

DOI: 10.22067/jsw.v0i0.9637

Authors

R. Daempanah^{* 1}; Gh.H. Haghnia¹; A. Alizadeh²; A.R. Karimi²

¹Department of Soil Science, College of Agriculture, Ferdowsi University of Mashhad

²College of Agriculture, Ferdowsi University of Mashhad

Abstract

Abstract
In the recent decades, application of geostatistic and remote sensing for mapping salinity and sodicity of surface soil and monitoring it's changes have been developed. The goal of this research was to compare the capability of geostatistic and remote sensing methods for mapping salinity-sodicity of soils in playa from sourthern Mah Valat county, in the Khorasan Razavi province. 88 samples of surface soil from depth of 0 to 10cm on the crossing lines of 1000×1000 meter grids were taken, and their EC, PH and SAR were measured. After normalizing variables, checking anisotropy and determining the parameters of variogram, the salinity and sodicity maps of soil were prepared using kriging method with an accuracy of 60 and 58 percent. the Results revealed that kriging had trimmed salinity and sodicity data because it had lowered the standard deviation. The IRS-P6 images were used to map salinity and sodicity maps. After pre-processing of images, PCA, salinity indexes of SI1, SI2, BI and NDSI were calculated and digital number of sampling points were extracted. By checking the correlation between variables and digital numbers of images, the SI1 that had the highest Pearson coefficient, was selected and salinity map of surface soil was prepared by the method of supervised classification. There was no correlation between images and sodicity data, and also between sodicity, salinity and pH data. Probability of extracting sodicity map by this method was evaluated poor. The result of this resarch also showed that for increasing accuracy in kriging maps more points should be taken, while in the remote sensing with less points while saving time and money, it’s possible to have more accurate maps.

Keywords: IRS, EC, Geostatistic, Salinity index, Superwised classification, Mah Valat