Vegetation is the basis for biodiversity conservation and sustainable development. In the Environmental Impact Assessment (EIA), which is the most direct and efficient policy measure to prevent degradation of nature, vegetation-related assessment has limitations as it is not based on quantitative and scientific methods. In addition, it focuses on the presence of protected species; hence, it does not take into account the role of vegetation as a habitat on a wide-area scale. As a way to overcome these limitations, this study aims to contribute to the quantification and advancement of future EIA on vegetation. Through the review of previous studies, core areas, connectivity, and vegetation condition were derived as the items to be dealt within the macroscopic aspect of vegetation impact assessment. Each item was spatially constructed using land cover maps and satellite imageries, and time series change analysis was performed. As a result, it was found that vegetation has been continuously deteriorating due to development in all aspects, and in particular, development adversely affects not only the inside of the project site but also the surrounding area. Although this study suggested the direction for improvement of the EIA in the vegetation sector based on data analysis, a more specific methodology needs to be established in order to apply it to the actual EIA process. By actively utilizing various environmental spatial data, the impact of the development on the natural ecosystem can be minimized.

본 연구의 목적은 1991년 위성 데이터와 2002년의 위성 데이터의 비교 분석을 통한 전남 완도의 난대림 식생 변화추이를 밝히는데 있다. 이를 위해 1991년 Landsat TM영상과 2002년 Landsat ETM 영상이 이용되었으며, 이미지 프로세싱은 ENVI를 이용하였다. GIS를 이용한 난대림 관리 시스템 개발은 Arc/Info와 ArcView를 이용하여 완성되었다. 영상 데이터의 촬영시기 차이와 TM 영상의 해상도(Resolution)가 정밀하지 못해 복잡한 지형특성을 지닌 곳에서는 정확한 변화추이를 파악하기는 불가능하였으나, 2002년 영상을 이용하여 분석된 완도의 상록활엽수림 면적은 약 2,027ha로 산출되었다. 상록활엽수림과 상록침엽수림은 11년 전에 비해 소폭으로 증가하였으나 낙엽활엽수림은 큰 폭으로 감소하였다. 식생분류를 위한 기준은 상록활엽수림, 상록침엽수림, 낙엽활엽수림, 기타로 구분하였으며 감독분류기법을 통해 식생이 분류되었다. 완도의 공간 데이터는 녹지자연도, 현존식생도, 산림토양도, 훼손유형도, 지형도, 토지소유현황도로 분류되었고, 이에 대한 속성 데이터 베이스는 Arc/Info와 ArcView를 이용하여 완성되었다. 데이터를 관리하고 필요한 정보를 색인 분석하는 사용자 GUI(Graphic User Interface)는 Avenue를 이용하여 개발되었다.

In this study, the yearly mean normalized difference vegetation index(YMNDVI) in Chungcheongnam-do was calculated using S10 NDVI data of the vegetation sensor for SPOT 4 and 5. Based on this calculation, statistical values such as mean value, standard deviation and coefficient of variation were determined. In addition, a comparative analysis was performed by calculating YMNDVI for cities and counties of Chungcheongnam-do. The YMNDVI of Chungcheongnam-do revealed a slight increase during 14 years between 1999 to 2012. However, it showed only a slight change within the range of 0.476 to 0.553, and no significant increase or decrease was noted. As a result, the highest YMNDVI was 0.553 at 2009, the lowest YMNDVI was 0.476 at 2001 and 2006. The mean value of YMNDVI in Chungcheongnam-do for 14 years turned out to be 0.502. As a result of the regional YMNDVI analysis, the highest YMNDVI region was Geumsan-gun, followed by Geryong-si, Cheongyang-gun and Gongjusi. The lowest YMNDVI region was Taean-gun, followed by Dangjin-si, Seosan-si and Seocheon-gun. An analysis of coefficient of variation in the research area showed that the mean value of Chungcheongnamdo was 4.2%, while the overall value was also not that high.

The western coast of South Korea is famous for its large and broad tidal lands. Nevertheless, land reclamation, which has been conducted on a large scale, such as Sihwa embankment construction project has accelerated coastal environmental changes in the embankment inland. For monitoring of environmental change, vegetation change detecting of the embankment inland were carried out and field survey data compared with Landsat TM, ETM+, IKONOS, and EOC satellite remotely sensed data. In order to utilize multi-temporal remotely sensed images effectively, all data set with pixel size were analyzed by same geometric correction method. To detect the tidal land vegetation change, the spectral characteristics and spatial resolution of Landsat TM and ETM+ images were analyzed by SMA(spectral mixture analysis). We obtained the 78.96% classification accuracy and Kappa index 0.2376 using March 2000 Landsat data. The SMA(spectral mixture analysis) results were considered with comparing of vegetation seasonal change detection method.