Using the Time Series Landsat Images to Detect the Land Use/Land Cover Changes in Dong Trieu District, Quang ninh Province, Vietnam

• Main Authors: Thi Thu Vu, 武氏秋
• Other Authors: Chaur-Tzuhn Chen
• Format: Others
• Language: en_US
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/20879354613578398053

碩士 === 國立屏東科技大學 === 土壤與水工程國際碩士學位學程 === 104 === Land use and land cover change are driven by human actions, particularly by population expansion and urbanization. As a consequence, it leads to heterogeneity in landscape and undermine environmental health as well. Therefore, this study was aimed at understanding land use and land cover change in Dong Trieu District - Quang Ninh Province - Vietnam. Four satellite imageries (TM 1990, 2000, 2010 and OLI/TIRS 2014) and demographic data were used for the study. Supervised image classification technique was applied to the four images and the after the image were spectrally and spatially enhanced. ERDAS imagine remotely sensed software was used for classification, and ArcGIS 10.1 was used to export the analysis images as maps. Google map, Historical land use map, and prior knowledge of the study area were used as referenced data for the classification. Six land use classes were used by the study, i.e. forest land, crop land, orchards, water body, built-up land, and barren land. This study used the spatial analysis tools available in Arc GIS 10.1 for the creation of change detection map. Finally, Fragstats software was used to evaluate the magnitude of fragmentation in this area. The study obtained a good result of some increased area and some decreased area results from all for images data of the four different dates, that were 1990 image, 2000 image, 2010 image, and 2014 image were merged as these inputs to produce four images of change detection results. Post classification analysis was used in order quantify the changes that took place during the period of the study. The findings revealed that forest had not much change in quantity but in quality, built-up areas had increased by 2.6 times between 1990 and 2014, orchards had increased by about 2 times, while the crop land, water body and barren land were decreased by 2.12%, 3.34%, and 11.21% respectively. Both RS and GIS techniques were fully utilized in this study, and the thesis had provided (from the analysis) a supporting data and good information that could be used for decision making in land use planning or urban planning. Fragstats software also presented the ability in landscape fragmentation evaluation and it can assist in studying landscape planning. Keywords: Landsat imageries, Land use change, Landscape ecology