Image Transformatiom and Geo-statistical Techniques to Enhance and Map Soil Features: an Application in Assessing Sedimentation Problem in Southern Nepal

A method is proposed to transform the red and near infrared data into a soil index in order to map soil features. The index maximises soil variation and helps improve soil feature mapping while it suppresses spectral response from vegetation cover. Geo-statistical analysis of fi eld data, on the other hand, helps understand spatial dependency pattern and map it, which may not be directly visible by remote sensing techniques. A case study in mapping sedimentation problem in the lowlands of Nepal is described. Yearly deposit of silt brought by irrigation water degrades soil physical condition and lowers fertility status, which results in eventual reduction of crop yield. Image transformation by means of band rotation helps generate soil index, which allows not only mapping areas affected by sedimentation but also separating areas with recent sediment deposition. Spatial variability study helps understand sediment deposition pattern and map the extent of sediment depth.