干旱区生态环境知识资源中心

During the last decades, Greater Cairo, Egypt, is increasing in population and in built-up extension. Some of the new buildings are informal, constructed in absence of government planning processes, and threaten the Heritage Cultural Site of the Giza Pyramids. In addition, the fertile land of the Nile floodplain is being urbanized despite the government’s building prohibition since the 1990s. Therefore, constant monitoring of construction activity is crucial in the rapidly changing environment of this area. Here, we present a data fusion approach that overcomes the limitations of single medium resolution sensor approaches, and also identifies areas in transition from desert to urban. We use multi-temporal multi-sensor supervised land use classification and include a new land use class for detecting undefined disturbances. Synthetic aperture radar ( SAR) data is combined with multi-spectral data for creating the land use land cover (LULC) maps using artificial neural networks (ANN). Specifically, ERS SAR data is combined with Landsat 5TM for 1998 and Envisat ASAR IMS with Landsat 7 ETM+ for 2004 and 2010. With this data fusion approach, it is measured an increase of 73% of Greater Cairo built-up extent from 1998 to 2010. Finally, we show the relationship between the aforementioned disturbances and the new built-up areas, detecting 26% of the total new built-up areas constructed from 1998 to 2010 where undefined disturbances were identified in previous land use maps. (C) 2016 Elsevier Masson SAS. All rights reserved.