| Aluminium industry is a very important fundamental industry forthe development of national economy. It is necessary that enoughavailable bauxite resources is needed to support the persistent and stabledevelopment of the aluminium industry. Therefor, it is essential toprospect for bauxite. At present, there are lots of methods being used toexplore the bauxite resources at home and abroad. According to differentgenetic types of bauxite, the prospecting methods are varies, includinggeological mapping, geomorphology, remote sensing geology,geophysics and so on. In this paper, it is mainly discussed a remotesensing(RS) mineralized information extraction method, which is usedin the study and explore the lateritic bauxite on Bolaven Plateau of Laos.Pakxong tenement in the southeastern Bolaven Plateau of Laos isour experimental area. The landsat ETM+ data is full used to study amineralized information extraction method in the research. Firstly, thegeological and remote sensing data are synthetically analyzed to makeclear the internal relationship between them. For example, we study thespectral characteristics of ion or ion group of the altered mineral, and thespectral characteristics and differences in the corresponding wave bands(ETM1～5,7) of the ETM+ RS image. On basis of these researches, aninformation extraction model of the enhanced mineralized and alteredinformation is formed. Secondly, the ETM+ data is pretreated, includingatmosphere correction, geometry correction, mosaic and so on. Thirdly,based on the enhanced mineralized information extraction model, afterseveral times of experimenting and analyzing, the best assembled wavebands are chosen and then a RS image is drawn out. In the RS image, themineralized information is enhanced greatly. The mineralizedinformation extracted from the RS image is checked by field geologicalsurvey and be proved that the extracted information is quite correspondto the condition of our tenement. After these, the RS image isprogressively processed with technique of supervise classification inorder to wipe off or decrease the interfered information of the extractedmineralized information. Meanwhile, the acreage of the mineralized areacan be calculated too. Here, a formula can be used to express this method, that is "Multi-data Analysis+Ratio+Principal ComponentAnalysis (PCA)+Classification", where, "Ratio+PCA", which is usedfor enhancing the mineralized information. In the study, "ETM5/1"and "PCA(1,4,5,7)" were selected, and "PC1(R)+ETM5/1(G)+PC3(B)"were chosen as the best assembled wave bands. Thus, an enhancedmineralized information pseudo-colour RS image is formed, and theinformation of bauxitization have been extracted too.The general conditions of stratum, structure, magmatic rocks areunderstood by field geological survey and by interpreting and analyzingof the RS image. Combining with the metallogenetic geological setting,a prospecting model for lateritic bauxite is formed. Based on thesuccessful extraction of the mineralized information in Pakxong workarea, this RS method is used to predict the probable bauxitization areason Bolaven Plateau, and 8 probable lateritic bauxite areas are predicted.Lastly, some measured resource calculation parameters of Pakxong wereused to infer the bauxite resources on Bolaven Plateau, and a total of1.877 billion tons of bauxite resources is gained. |
PDF Full Text Request