Monitoring LULC Changes in El-Fayoum Governorate using Support Vector Machine.
Paper ID : 1094-ICRSSSA-FULL (R2)
Authors
Islam Atef Fouad *1, Wael Mohamed Ahmed2, Ramadan H. Abdel-Maguid1
1Civil Engineering Department, Faculty of Engineering, Fayoum University, Fayoum 63514, Egypt.
2Geomatics Engineering Lab, Civil Engineering Department, Faculty of Engineering, Cairo University, Giza 12316, Egypt.
Abstract
Since the 1980s, El-Fayoum has witnessed several changes in land use. Images from Landsat satellites were used to examine the changes in land use/cover in EL-Fayoum from 2000 to 2020. For assessing quantitative data and processing satel-lite images for this study area's assessment of land use change, Google Earth En-gine (GEE) and ARCGIS Pro were utilized. GEE makes the processing and pre-processing required for satellite images fast and easy. The supervised land use classification process was compared between using a support vector machine al-gorithm (SVM) and maximum likelihood (MLH). SVM was better in accuracy than MLH, which was a respectable result for monitoring changes. It was dis-covered that within 20 years, respectively, 94.22 km2 and 6.39 km2 of deteriorat-ed agricultural land had been converted into urban areas and desert regions. An-other 143.72 km2 and 111.96 km2 of desert land were transformed into agricul-tural land and urban areas, respectively. The area of the water body decreased from 337.76 km2 to 315.44 km2 with an annual change rate of -1.11%. The ur-ban area increased from 200.70 km2 to 350.34 km2 with an annual change rate of 7.48%. The findings of this study will be useful in organizing and putting into practice crucial management choices in order to preserve El-Fayoum's biodiversi-ty.
Keywords
Google Earth Engine; Change detection; Support vector machine; remote sensing, GIS.
Status: Accepted (Oral Presentation)