Various human activities such as the Urbanization and Industrialization are estimated the main factors to
pollute the stream. Now days, numerous studies are carried out for managing non-point sources which have un-effect
on water quality of streams by land-use and livestock. In case of Korea, a matter of concern that the management
of Pal dang reservoir - the main water resources of the national capital region - has been occurring. Especially,
large-scale constructed-wetlands are planned and constructed at the end of Kyoung-an stream. Additionally a lot of
sewage treatment plants are newly installed and extended in this watershed. According to these efforts, water quality
of Kyoung-an stream is predicted that would be improved. But the more detail and scientific analysis should be
carried out for the water quality improvement, because, existing water quality improvement projects are not involved
to analyze root of water quality deterioration and improvement plans. Therefore, this study aims to select suitable
areas for constructed-wetlands and to calculate size of the constructed-wetlands for water quality improvement in
Kyoung-an stream through the geographical pollutant distribution analysis and land-use pattern analysis by high
resolution satellite image and suitable area analysis of constructed-wetlands by GIS(Geographic information system).
The progress of this study is to select maximum pollutant loaded area by geographical analysis based on wate① r
quality data, ② to analyze land-use patterns using high resolution satellite image, ③ to select suitable areas of
constructed-wetlands, ④ to calculate area and volume of chosen constructed-wetlands using GIS. Basically, sizes of
constructed-wetlands are induced through the constructed-wetlands design index based on treatment ratio(provided by
Korea Water Resources Corporation). As a result of this study, two areas are selected to construct
constructed-wetlands. One of the area was 127,586㎡ near by Yong-in sewage treatment plant, and the other area
was 1,647㎡ near by Ju-buk stream and Dae-dae stream.