A Prediction Model for Soil Salinity Using its Indicators: A Case Study in Southern Iran

South of the Zagros belt, the entire land of Southern Iran faces problems arising out of various types of land degradation of which soil salinity forms a major type. The Mond river basin, located centrally to this zone, has been selected as a test area to develop a statistical model for predicting the salinity of soil using different indicators of soil salinity. The soil salinity data were taken at 49 different samples in the study area. The data as indicators of soil salinity have been gathered from the records and reports published by the different departments of the Ministries of Agriculture, Defence and Energy of Iran. The GIS analysis of various indicators and salinity of soil samples considered proved useful for understanding their relationship in a statistical software. In the present study, the relations between the soil salinity and the indicators of soil salinity have been found statistically in the software of SPSS. To find a regression equation for soil salinity, max EC in 1 m depth of soil has been considered as dependent variable while the indicators of soil salinity including soil texture, water table, dry index, slope, index of efficacy of surface geology (ESG) and groundwater quality are considered as independent variables. For this purpose, the regression equations for two methods of 'enter' and 'stepwise' in software of SPSS have been established. The linear regression equations define the variations of the soil salinity depending on the indicators and also give an idea about the levels of relations. The results obtained show that the relations between the soil salinity and the indicators especially groundwater data do exist.