Modelling the Behavior of DVI and IPVI Vegetation Indices Using Multi-Temporal Remotely Sensed Data

Abstract

Remote sensing techniques are widely used to detect and analyze land cover changes due to their accuracy and cost-effectiveness. Among the various spectral indices derived from the satellite data Difference Vegetation Index (DVI) and Infrared Percentage Vegetation Index (IPVI) vegetation indices applied to model the behavior of the indices in the study of suburb ecosystem vegetation cover over twenty years. To achieve the aim of the study two objectives were formulated; detect Spatial-temporal variations in urban vegetation and how suitable the selected algorithms to study urban ecosystem vegetation. The study area is a rapidly developing area consists of several suburbs including Battaramulla, Malabe, and Kaduwela, Sri Lanka. The study used Landsat data and pre-processing, processing, geometric and atmospheric corrections were performed using ERDAS imagine mapping software and all the mappings were carried out via Arc GIS software. The results show Infrared Percentage Vegetation Index (IPVI) algorithm as the most suitable vegetation index to study suburb ecosystem vegetation than Difference Vegetation Index (DVI) in the study area. Therefore, the study recommends IPVI than DVI to study ecosystem vegetation in sub-urban areas.