Object-based image analysis for the impact of sewage pollution in Malad Creek, Mumbai, India.

Abstract

Today, object-based image analysis provides an option for integrating spatial information beyond conventional pixel-based classifications for high-resolution imagery. Due to its rare applicability in pollution assessment, an attempt has been made to assess the spatial extent of sewage pollution in Malad Creek, Mumbai, India. Based on multiresolution segmentation of an IRS P6 (LISS IV) image and the Normalized Difference Turbidity Index (NDTI), the various water quality regions in the creek were classified. The existing literature implies that the reflectance of turbid water is similar to that of bare soil which gives positive NDTI values. In contrast to this, negative values of NDTI are observed in the present study due to the presence of organic matter which absorbs light and imparts turbidity, which is supported by the significant correlation between NDTI and turbidity. A strong relationship is observed between turbidity and water quality parameters, implying the impact of organic matter through discharges of sewage in the creek. Based on the classified regions and the water quality parameters, the extent of pollution was ranked as high, moderate, low and least. The methodology developed in the present study was successfully applied on an IKONOS image for the same study area but a different time frame. The approach will help in impact assessment of sewage pollution and its spatial extent in other water bodies.