Using a time series of satellite imagery to detect land use and land cover changes in the Atlanta, Georgia metropolitan area

Abstract

A time series of Landsat MSS and TM images was used to extract land use/cover change data of the Atlanta, Georgia metropolitan area in the United States over the past 25 years as a component of Project ATLANTA (ATlanta Land-use ANalysis: Temperature and Air-quality). ATLANTA is funded by NASA EOS Interdisciplinary Science (IDS) program, which has the objective of modelling the impact of land use/cover change on temperature and air quality in Atlanta. This paper describes a suite of techniques that have been used to develop an operational approach, which will ensure high accuracy and compatibility in image classification from the satellite images of different resolutions and varying quality. These techniques include radiometric normalization to establish a common radiometric response among multi-date/multi-sensor data, an unsupervised image classification approach using image clustering and cluster labelling, a GIS-based image spatial reclassification procedure to deal with classification errors caused by spectral confusion, and post-classification comparison with GIS overlay to map the spatial dynamics of land use/cover change. The loss of forest and urban sprawl have been revealed by the analysis as the major problems of Atlanta's accelerated urban development.