2004 Denver Annual Meeting (November 7–10, 2004)

Paper No. 17
Presentation Time: 1:30 PM-5:30 PM


SHI, Peichang1, CHENG, Songlin2 and SMITH, Steven B.1, (1)Environmental Sciences Ph.D. Program, Wright State Univ, 3640 Col. Glenn Hwy, Dayton, OH 45435, (2)Department of Geological Sciences, Wright State Univ, 3640 Col Glenn Hwy, Dayton, OH 45435, songlin.cheng@wright.edu

The first documented coal mining in Ohio was in 1800. For the first 150 years of Ohio coal mining, over 3.4 billion tons of coal has been mined. Consequently, today there are over 6,000 known abandoned underground mines in Ohio and estimates vary from 2000 to 4000 more unknown underground mine. Lakes and streams affected by coal mines have very little biological activity due to low pH and high heavy metal concentration in the water. Open coal waste piles continue to generate acidity and acidify surface water. The purpose of this study is to study the feasibility of using GIS and Remote Sensing to identify acid lakes. The initial attempt to identify lakes by image classification using LANDSAT 7 imageries didn’t produce a unique cluster of lakes. Digital Line Graph (DLG) of lakes was then used to isolate lakes to produce lake layer. It was anticipated that a lack of phytoplankton in acid lakes should have uniformly low Normalized Difference Vegetation Index (NDVI), while normal lakes should show a high value in the summer and low in the winter. IDRISI32 was used to generate NDVI from LANDSAT 7 bands 3 and 4 of the study site between March and December. The limitation of suitable data made it necessary to use LANDSAT 7 data taken between 1999 and 2002. Boolean operation using lake layer to multiply NDVI isolate the NDVI of lakes from non-lake areas. The result clearly shows that NDVI of normal lakes are high during the growing season (> 0.4) and approaching 0 during the winter, while NDVI of acid lakes is about -0.2 in each month. It is concluded that this approach can effectively differentiate normal lakes from the acid lakes, especially NDVI during the growing season.