"It must comprehensively analyze the satellite photographs and aerial photographs in combination with other data and, on this basis, clarify three-dimensionally the geological formation and distribution of underground resources to the depths."
In general, the hydrothermal alteration zones might be associated with the gold mineralization. This plays an important role at the early exploration step of the mineral prospecting. The detection of hydrothermal alteration zone by the multispectral satellite image data have been widely used in the prospecting of the ore deposit because of its labour-saving and low cost. The host rocks that contain ore deposits or enriched in ore minerals of hydrothermal origin always carries the end-products of interaction with the hydrothermal fluids that could change the ore minerals and chemical composition of the rocks and cause the deposition of ore and related hydrothermal minerals. Hydrothermal alteration zones are commonly associated with the propylitic minerals (chlorite, epidote and calcite), argillic minerals (kaolinite, alunite, and montmorillonite), phyllic alteration minerals (sericite, illite), and the potassic alteration minerals (K-feldspar, muscovite). Also the weathered gossans which have anomalies of Au, Cu contain the iron oxides (hematite, goethite, and jarosite).
This leads to many studies that have been carried out for identifying the altered minerals from the multispectral satellite image. Some researchers clarified that most of the alteration minerals have diagnostic absorption features in the SWIR bands, which offer potential for mineral identification (such as gold and copper) through remote sensing studies. Also the other researchers suggested the research result which the ferrous ion bearing minerals (goethite, hematite, jarosite) and hydroxyl bearing minerals (clay minerals and micas) clearly show diagnostic absorption features of the mineral assemblages associated with the hydrothermal alteration zones in the visible, near infrared, short wave infrared regions from Landsat TM/ETM+ image data. Based on the theory above, many earlier researchers indicated various methods for the mineral prospecting to successfully identify both iron oxides and clay minerals within the hydrothermal alteration zones in hydrothermal and porphyric origin deposits using the different kinds of image prospecting methods such as the false color composite images, ratio image, PCA etc.
Landsat 8 OLI basically coincides with Landsat 7 ETM+ in both the spatial and spectral characteristics. Compared to Landsat 7 ETM+, the bandwidth of band 5 in OLI sensor is adjusted in the range of 0.845-0.885㎛. The water vapor absorption in the vicinity of 0.825㎛ is eliminated. The bandwidth of band 8 (panchromatic band) is also narrower than that of Landsat ETM+, therefore, it allows vegetation cover area and non-vegetation area can be distinguished better. OLI has two more bands than that of ETM+. These are band 1 (0.433-0.453㎛) which is used in the coast observation and band 9 (1.360-1.390㎛) which is used in the cirrus observation. Also all the OLI and TIRS spectral bands are stored as geo-referenced 16 bit digital numbers. Thus we used the Landsat 8 OLI image data to detect the hydrothermal alteration zones associated with gold mineralization.
We reduced the effect of vegetation which impacted on the extraction of mineralization alteration zone, using the reduction method of vegetation effect based on the linear spectral unmixing theory. Then we extracted the hydrothermal alteration zone related with the gold mineralization and indicated the direction of gold prospecting.
In order to eliminate the effect of vegetation from the satellite image data, we first calculated Normalized Difference Vegetation Index (NDVI), then extracted the end member spectra of vegetation in the area with more than 0.75 of NDVI. Next, using the matched filtering method based on the linear spectral unmixing theory, the vegetation sub-pixel abundance was estimated at each pixel of the multispectral image and subtracted from the spectrum of each pixel. After this, spectral effects of vegetation were removed.
For mapping iron oxides and clay minerals from satellite image data, band 2, 4, 5, 6 were selected, then the Principal Component Analysis (PCA) was applied on two datasets. To draw a mapping of the iron oxides and clay minerals from PCA and four OLI (2, 5, 6, 7), we analyzed the loadings on each band, and determined the PC 4 as iron oxides image and clay mineral image respectively.
False color composite image was generated by composite of the hydroxyl image in red channel, the iron oxide image in green channel and the average of these two images in blue channel, All the hydrothermal alteration zones are shown brighter color and verified by field survey. The newly detected alteration zones from the satellite image data included the distribution area of the auriferous quartz veins.
Our study result was published in the international academic journal "Journal of the Indian Society of Remote Sensing" under the title of "Detection of hydrothermal alteration zones using Landsat 8 OLI image: A case study of gold prospecting in Nyongwon area, DPR Korea"(https://doi.org/10.1007//s12524-021-01385-8).