REMOTE SENSING DETECTION FOR SUBSIDENCE-RESULTED WATER BODY AND 
SOLID-WASTE DUMP IN COAL MINE: YANZHOU BEING A CASE 
MA Baodong 3 WU Lixin a,b LIU Shanjun 3 
institute for Geoinformatics & Digital Mine Research, Northeastern University, Shenyang 110004, China, - 
mabaodong_rs@ 126.com 
institute for GIS/RS/GPS & Subsidence Research, China University of Mining & Technology, Beijing 100083, China, 
- awulixin@263.net 
KEY WORDS: Land Cover, Change Detection, Coal mining, Satellite Image, NDWI, Infrared 
ABSTRACT: 
Taking LandSat TM/ETM+ images as information source, this paper used the Normalized Difference Water Index (NDWI) to extract 
mining subsidence-resulted surface water body information in Yanzhou coal mining area. It was found that the extracted water body 
information would be mixed with solid-waste dumps. Taking DN 40 in band 5 as threshold value, the water bodies and solid-waste 
dumps were identified accurately. In addition, DN 80 in band 3 could be used as threshold value to identify shallow water and deep 
water. Furthermore, the results based on NDWI-threshold method were cross validated by IR band, and the changing trend of water 
bodies and solid-waste dumps was detected in Yanzhou coal mining area in the past 15 years. 
1. INTRODUCTION 
Satellite remote sensing plays an important role in 
environmental monitoring for mining area. It has the following 
advantages: 1) images cover large areas on the ground; 2) it is 
not time consuming but has sufficient temporal frequency; 3) 
prices per square kilometer are generally lower than in-situ 
investigation and monitoring. 
A number of papers focus on monitoring the environment of 
mining area by using of various remote sensing data. Mine 
wastes and lands affected by selenium-rich water runoff, in 
southeast Idaho, were mapped and analyzed by AVIRIS 
imagery and digital elevation data (John et al., 2003). With the 
help of remote sensing technology, in the St. Austell China clay 
(kaolin) region, Cornwall, UK, the waste tips were identified 
and classified and its spatial distributions were mapped (Richard 
et al., 2004). In Kailuan coal mining area, north China, 
multi-temporal ERS1/2 SLC SAR data were used for the 
monitoring and the dynamical analysis of surface subsidence 
(WU Li-xin et al., 2005). In the Raniganj Coalbelt, India, 
temperature anomalies caused by coal-fires were identified (R.S. 
Chatteijee, 2006). The numerous studies indicate that remote 
sensing is an effective method for environmental investigation 
and monitoring in mining area. 
SELECTED MINING AREA AND SATELLITE 
DATA 
Yanzhou coal mining area, which lies in the southwest plain of 
Shandong province (Fig.l), has an annual yield of more than 3 
000xl0 4 tons raw coal. It includes Xinglongzhuang, Dongtan, 
Baodian, Nantun, Beisu, Yangcun, Jierkuang coal mine, etc. 
Around the mining area, there is large area of fertile farmland. 
The massive coal mining is making a strong environmental and 
ecological impact on surface environment. Mining 
subsidence-resulted surface water bodies and solid-waste dumps, 
which take a large area of farmland and become the two main 
aspects of the mining impact. 
LandSat TM/ETM+ images may be the most frequently used 
data for monitoring land cover change. The spatial resolution of 
the visible and reflected IR bands is 30 m, and that of the 
thermal IR band 6 is 120 m for TM and 60 m for ETM+. This 
paper selected a group of multi-temporal LandSat TM/ETM+ 
images, acquired in September 19, 1987 and May 31, 2002 
respectively (http://glcfapp.umiacs.umd.edu), as the information 
source to detect the environment changes during the past 15 
years.