Remote sensing-based identification and mapping of salinised irrigated soils: a case study along the Orange River between Upington and Keimoes, South Africa

摘要

Salinisation is a major environmental hazard that limits soil productivity. It may be categorized as primary or secondary soil salinisation. Owing to the low costs and time saving associated with the use of remote sensing, remote sensing data was used in this study to identify and map salinised irrigated land along the Orange River between Upington and Keimoes, Northern Cape Province, South Africa. This study aimed to evaluate the potential of high resolution digital aerial imagery in identifying salinised cultivated land. The methods used were visually identifying stressed crops on near infrared and NDVI images, and performing unsupervised image classification. Soil samples were collected and analysed for salinity to validate the results. Visual image analysis had an overall accuracy of up to 70% while unsupervised classification achieved an overall accuracy of only 50%. Neither method could differentiate salt induced stress from other forms of crop stress.