Application of knowledge-based decision tree classification method to monitoring ecological environment in mining areas based on the multi-temporal Landsat TM(ETM) images: a case study at Daye, Hubei, China

This paper presents a case study of Daye, Hubei, China, to trace mining activities and related environment changes during the past 10 years, with an emphasis on land cover changes. Two sets of satellite data have been used: TM and ETM+ image data. A multi-temporal dataset consisting of two Land sat 5 Thematic Mapper (TM) images and one Enhanced Thematic Mapper Plus (ETM+) image in 1986, 1994 and 2002 have been used to compare the land cover changes of the Daye area, Hubei Province, China. Combined bands method and iron oxide index and the NDVI index method have been used to investigate the spectrum character and the space character of the different ground objects. The knowledge-based decision tree classification method has been used to get highly accurate classification result from the TM and ETM+ image data. The results of change detection show that quality of whole water body was still bad, although the water quality has been improved in some areas. Vegetation shows that degradation trend occurs especially in those areas close to the mining areas, large areas of wood land and plantations are reduced, the increasing bare areas appear and the reclamation percentage of the abandoned mining is only 20% from 1986 to 2002. The ecological environment in the study area may become worse unless the efficient management of mining and effective eco-environment protection are carried out instantly.

[1]  Dong Guang Primary Study on the Multi-Layer Remote Sensing Information Extraction of Desertification Land Types by Using Decision Tree Technology , 2000 .

[2]  Sholom M. Weiss,et al.  Computer Systems That Learn , 1990 .

[3]  Li Wenjun,et al.  Primary study on the multi-layer remote sensing information extraction of desertification land types by using decision tree technology , 2002, IEEE International Geoscience and Remote Sensing Symposium.

[4]  Philip H. Swain,et al.  Purdue e-Pubs , 2022 .

[5]  Li Jing Classification Methods of Land Use/Cover Based on Remote Sensing Technology , 2004 .

[6]  Alan H. Strahler,et al.  Maximizing land cover classification accuracies produced by decision trees at continental to global scales , 1999, IEEE Trans. Geosci. Remote. Sens..

[7]  A. Strahler,et al.  Application of the MODIS global supervised classification model to vegetation and land cover mapping of Central America , 2000 .

[8]  R. Almeida-Filho,et al.  Digital processing of a Landsat-TM time series for mapping and monitoring degraded areas caused by independent gold miners, Roraima State, Brazilian Amazon , 2002 .

[9]  R. DeFries,et al.  Classification trees: an alternative to traditional land cover classifiers , 1996 .

[10]  Alan H. Strahler,et al.  Global land cover mapping from MODIS: algorithms and early results , 2002 .

[11]  H. Schmidt,et al.  Multitemporal analysis of satellite data and their use in the monitoring of the environmental impacts of open cast lignite mining areas in Eastern Germany , 1998 .

[12]  J. Townshend,et al.  Global land cover classi(cid:142) cation at 1 km spatial resolution using a classi(cid:142) cation tree approach , 2004 .

[13]  G. Venkataraman,et al.  Open cast mine monitoring and environmental impact studies through remote sensing ‐ a case study from Goa, India , 1997 .

[14]  P. Mather,et al.  Classification Methods for Remotely Sensed Data , 2001 .

[15]  David A. Landgrebe,et al.  A survey of decision tree classifier methodology , 1991, IEEE Trans. Syst. Man Cybern..

[16]  J. Townshend,et al.  Global land cover classifications at 8 km spatial resolution: The use of training data derived from Landsat imagery in decision tree classifiers , 1998 .