Remote sensing-based identification and mapping of salinised irrigated land between Upington and Keimoes along the lower Orange River, South Africa

[摘要] Salinisation is a major environmental hazard that reduces agricultural yields anddegrades arable land. Two main categories of salinisation are: primary and secondarysoil salinisation. While primary soil salinisation is caused by natural processes,secondary soil salinisation is caused by human factors. Incorrect irrigation practicesare the major contributor to secondary soil salinisation. Because of low costs and lesstime that is associated with the use of remote sensing techniques, remote sensing datais used in this study to identify and map salinised irrigated land between Upington andKeimoes, Northern Cape Province, in South Africa.The aim of this study is to evaluate the potential of digital aerial imagery inidentifying salinised cultivated land. Two methods were used to realize this aim. Thefirst method involved visually identifying salinised areas on NIR, and NDVI imagesand then digitizing them onscreen. In the second method, digital RGB mosaicked,stacked, and NDVI images were subjected to unsupervised image classification toidentify salinised land. Soil samples randomly selected and analyzed for salinity wereused to validate the results obtained from the analysis of aerial photographs.Both techniques had difficulties in identifying salinised land because of their inabilityto differentiate salt induced stress from other forms of stress. Visual image analysiswas relatively successful in identifying salinised land than unsupervised imageclassification. Visual image analysis correctly identified about 55% of salinised landwhile only about 25% was identified by unsupervised classification. The twotechniques predict that an average of about 10% of irrigated land is affected bysalinisation in the study area.This study found that although visual analysis was time consuming and cannotdifferentiate salt induced stress from other forms; it is fairly possible to identify areasof crop stress using digital aerial imagery. Unsupervised classification was notsuccessful in identifying areas of crop stress.

[发布机构] Stellenbosch University