ESTIMATING LAND USE/COVER OF FOSHAN CITY IN SOUTHERN CHINA WITH LANDSAT REMOTE SENSING IMAGERY FOR FLOOD MODELLING

References: 17th International Multidisciplinary Scientific GeoConference SGEM 2017, www.sgem.org, SGEM2017 Conference Proceedings, ISBN 978-619-7408-03-4 / ISSN 1314-2704, 29 June - 5 July, 2017, Vol. 17, Issue 23, 79-86 pp, DOI: 10.5593/sgem2017/23/S10.011

ABSTRACT

The most direct change of a watershed induced by urbanization is the land use/cover (LUC) change, which will change the flood response of the watershed. To study this effect, long series of LUC change at watershed scale is needed to be prepared. A watershed is usually a large domain even it is at small scale considering its fine spatial resolution, so the traditional way of field surveying is not feasible for mapping LUC at watershed scale. Development of satellite remote sensing technology provides inexpensive way for estimating LUC change at large domain. Landsat series satellite imageries are the most widely used imageries for estimating LUC changes at watershed scale due to its large quantities of archived imageries and easy accessibility. In this study the Landsat remote sensing imageries from 1987 to 2016 at an average 3 year interval were acquired for estimating the LUC changes of Foshan City in southern China, a newly developed metropolitan in the past 3 decades. SVM classification algorithm has been employed to estimate the LUCs of Foshan City automatically first, and then post-processed by manual interpretation. A dataset of LUCs of Foshan City at 12 time stages was finally prepared with good accuracy for hydrological modelling to study the flood response changes. Based on the result of this dataset, the LUC change pattern of Foshan City was also analyzed.

Keywords: urbanization, remote sensing, land use/cover, support vector machine, flood modelling