APPLICATION OF MULTI-TEMPORAL TM (ETM+) IMAGE IN MONITORING 
MINING ACTIVITIES AND RELATED ENVIRONMENT CHANGES: A CASE STUDY 
AT DAYE, HUBEI, CHINA 
Shiyong YU Zhihua CHEN 2 , Yanxin WANG 2 
(l.The department of well-logging and remote sensing technology, the research institute of petroleum exploration & development 
CNPC,Beijing 100083, P. R. China; 
2. School of Environmental Studies, China University of Geosciences, Wuhan 430074, P. R. China) 
Keywords: Multi-sensor; Remote sensing; Mining area; Eco-environmental monitoring; Fusion; 
Abstract: 
This paper presents a case study of Daye, Hubei, China, to trace the mining activities and related environment changes during the 
past 10 years, with an emphasis on land cover changes. The TM (ETM+) satellite data has been used in this case study. A 
multi-temporal dataset consisting of two Land sat 5 Thematic Mapper (TM) images and one Enhanced Thematic Mapper Plus 
(ETM+) image including 1986, 1994 and 2002 have been used to compare the land cover changes of the Daye area, Hubei Province, 
China. According to the study, the conclusion can be drawn the ecological environment in the study area may become worse unless 
the efficient management of mining business and effective eco-environment protection in mining area are carried out instantly. 
1. INTRODUCTION 
Remote sensing is the science of acquiring, processing, and 
interpreting images and related data, acquired from aircraft and 
satellites that record the interaction between matter and 
electromagnetic energy (Sabins, 1997). Mine exploitation of 
the Human impacts on the Earth system is considerable. The 
irrational mine exploitation leads to a series of ecology 
problems, for example, the vegetation degenerate, clean water 
contaminated, moreover, the geologic hazards can be induced 
more often than before such as the coast, block glacier and so 
on. All these can be a hamper to the national economy 
sustainable development (Tan Yongji, 2000). 
Until recently, the vast majority of mine environment 
investigation have focused on the qualitative analysis or half 
quantitative analysis which used the combined bands map, 
index change map and the land cover map to describe the 
changes of the earth surface, some successful quantitative 
analysis are mainly used in isolated local mine (Lei Liqing et 
al., 2002; F.Llorens et al., 2000; H.Schmidt &C.Glaesser, 1998; 
Raimundo Almeida-Filho&Yosio E.Shimabukuro., 2002; 
Venkataraman G. et al., 1997). On the other hand, some 
research integrates the remote sensing data with the other data 
(such as the soil texture and the water quality).Although the 
result is exact, the work need much time and manpower. 
(Venkataraman G. et al., 1997; Christian Fischer, Wolfgang 
Busch., 2002) 
Considering the scattered mine and the much kinds of ore 
deposit, this paper utilizes the multi-temporal satellite image 
data. The knowledge-based decision tree classification method 
was used to get the highly accurate classification result with the 
TM and ETM+ image data. 
2. METHODS 
2.1 Data and pre-processing 
The TM (ETM+) are derived from satellite images that were 
taken at three different times: July 30, 1986(TM), November 7, 
1994 (TM), September 4, 2002 (ETM+). These three images 
are chosen in consideration of the vegetation changes in the 
region of the Daye region. We use the ERDAS IMAGINE 8.7 
software to process and interpret the satellite images. The other 
data such as relief map have accomplished digitization map by 
the ArcGis 8.3 software, mine enterprise statistic data have 
been projected and corrected by MAPGIS6.3. In order to make 
the remote sensing data match the data selected from the data 
besides the remote sensing (such as relief map, mine enterprise 
annual repot and so on) ,The remote sensing image data must 
be corrected according to 1/50000 relief map. The correct 
points are selected on the remote sensing image and relief map 
respectively, the principle of selection is that the points, such as 
the cross of the roads, ends of the railway and the dam, must be 
discriminative, clear and typical. The binary quadratic 
polynomial is selected in this paper. 
The corrected image matches the 1/5 relief map exactly, the 
projection mode is Gauss Kruger, the result of correction 
answers for the total residual is less than 1 ,the error is less than 
one image cell. The corrected image data accord with research 
precision. 
2.2 Knowledge-based decision tree classification method