MONITORING AND PREDICTING URBAN GROWTH IN TAOYUAN CITY, TAIWAN

ABSTRACT: Previous studies have shown that urban growth can be related to the concentration of population, the development of transportation network, the increase of local income and the control of government policies. Land use changes referred to urban development are commonly related to anthropogenic causes. Meanwhile, physical environmental factors such as climate and topography could also guide these changes. The land use survey data, however, collected for assessing urban development is not always available. For recent decades, many researches have shown that satellite remote sensing data is capable of providing accurate and timely geospatial information to describe changes in land cover and urban shape. The study area, Taoyuan City, is the sixth and new major city of Taiwan. The city has the highest population growth rate in Taiwan from 1984 to 2014 and experienced a significant increase of urban land, due to the activation of High Speed Railway in 2007. In addition, the Taoyuan Aerotropolis project has been activated since 2014. Therefore, a rapid development of the city could be expected in near future. The study aims to assess the possible urban growth of Taoyuan City in the future. Specifically, this study uses Landsat data, including Landsat TM and OLI to analyze the historical urban land development of Taoyuan City, in 1984, 1995, 2006 and 2014. First, by applying support vector machine (SVM) classifier, different land cover types, including built-up, farm, vegetation, water and barren, have been classified and mapped. Second, this study quantified the land cover change and analyzed the transitions of land cover over the past 30 years. Third, based on historical urba