The eco-environment of the arid region in West China is quite feeble.It is of great practical significance to accurately and automatically extract salinized soil information to the monitoring of soil salinization and the study on dynamic variation of soil salinization.Remote sensing(RS) technology demonstrates a number of advantages in studying soil salinization in arid area.But how to extract salinized soil information accurately from RS images is the basis of the study.An arid area typical of the Wigan-Kuqa River Delta Oasis in the north of the Tarim River Basin of Xinjiang,China,was selected as an example for the study.Seven-spectrum-band Enhanced Thematic Mapper-plus(ETM+) images dated August 1,2001 were used against the data obtained from field investigation of the soil features of the area,and analysis of typical soil RS soil information,to extract NDVI(Normalized Difference Vegetation Index) and PC3(third principal component),which were specified as major characteristic variables for information extraction of non-salinized soil and salinized soils,respectively.And MNDWI(Modified Normalised Difference Water Index),TM1 and TM7 were cited as supplementary characteristic variables.An information extraction model,using the decision-tree classification method was established and applied to classification of RS images.Results show that the RS information characteristics-based automatic information extraction of salinized soils is feasible and capable of yielding information higher in identification accuracy.