Abstract: Soil salinity is a critical problem in many arid and irrigated agricultural areas. It reduces soil quality and limits the growing of crops. Monitoring soil salinity is important for controlling this problem. During the last decades, many researches aim to develop methods for assessing, mapping, and monitoring salt-affected soils using remote sensing data in combination with field measurements. The methods were classified as visual interpretation of satellite imagery, digital analysis of multispectral remote sensing data, quantitative analysis of hyperspectral imagery and integrated modeling. The main objective of this study is to assess the usefulness of the thermal infrared (TIR)emissivity for predicting salt content. For this purpose, soil samples were collected from the field for laboratory determination in this study. And spectral data was acquired from Fourier Transform spectrometer (102F), predicting model was developed using a Regression Model. The spectral results show that different salt contents in soil exhibited different emissivity. The emissivity were extremely sensitive to salt content in the 8.5-9.5 μm region, with a high correlation value (0.8 in the 9.259-9.271 μm region), holding a great promise for prediction of salt content; the Stepwise Multiple Regression Mode performed well (R²=0.899, RMSE=1.734) based on the first derivative spectral pre-treatment. The results of this study indicate that infrared thermal data provide a useful tool for predicting salt content in soil that commonly occurs in aid region.

Key words: salt content, Thermal Infrared Remote Sensing, Soil emissivity spectra, Stepwise multiple regression(SMR)