Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
J. Environ. Treat. Tech.  
ISSN: 2309-1185  
Journal web link: http://www.jett.dormaj.com  
Detection of Hydrothermal Alteration Zones  
using Image Processing Techniques, Chahr  
Gonbad, Iran  
1
2
3
Abbas Ali Nouri , Rasoul Sharifi Najafabadi , Mohammad Ali Nezammahalleh *  
PhD Candidate in geomorphology, Kharazmi University, Tehran, Iran  
Assistant Professor of Geomorphology, Farhangian University, Tehran, Iran  
PhD in geomorphology, University of Tehran, Tehran, Iran  
Received: 27/09/2019  
Accepted: 12/12/2019  
Published: 20/02/2020  
Abstract  
Use of satellite images to detect hydrothermal alteration zones can be helpful for efficient mineral explorations. Remote Sensing  
RS) techniques make it possible to save cost and time for accurate primary explorations. The purpose of this research is to use the  
(
image processing techniques to detect porphyry copper minerals through the index of hydrothermal alteration mapping in Chahr  
Gonbad, Sirjan, Iran. We have obtained Landsat imagery data of Enhanced Thematic Mapper (ETM) and Operational Land Imager  
(
OLI) for the analysis. For enhancement and detection of the alteration zones, we have applied useful RS methods including image  
difference and band color composite, band ratios, Principal Component Analysis (PCA), Crosta, and Matched Filtered. The results  
discovered argillite and propylitic alteration zones through iron-oxide and hydroxyl minerals in the study area. The image  
processing has revealed that the detected alteration minerals have relation with mineralization of porphyry copper. The zones are  
mainly developed in a northwest-southeast orientation and mostly concentrated in center and south of the Chahr Gonbad.  
Keywords: Porphyry copper, Band ratio, PCA, Crosta, hydrothermal alteration mapping, Chahr Gonbad  
Introduction1  
Ore deposits are geologic 3-dimensional features that  
1
include unusual concentration of one or more elements. They  
are predominantly differentiated from surrounding areas.  
Exploration of mineralization is based upon use of  
exceptional conditions in 4 factors including elements,  
mineralization, petrology, and structure. To make this work,  
in exploration simplified models of mineralization and  
metallization are used in some steps.  
Formation of mineral reserves is  
a complicated  
geological process (1). Mining engineers and geologists  
have classified a variety of processes forming mineral  
materials (2). Hence, the evidence of a variety of minerals is  
greatly acknowledged and applied in mining explorations  
(
3). Today, the science of remote sensing (4) by many  
helpful capabilities in processing of satellite images is very  
important in earth science studies (5-8) including mining  
explorations, recognition of rock material in geology,  
detection of faults, mapping, and creating Digital Elevation  
Model (DEM). The most important advantage of this  
technique is to have access to essential information in a  
minimum of time. Since satellite images have spectral data  
in all ranges of electromagnetic spectrum, they provide  
useful information about the nature of land surface features.  
Another advantage of the data is appropriate repetition of  
imaging that make it possible to investigate geodynamic  
phenomena (9). The fundamental of these data is based on  
remotely measuring reflection of the earth features. Hence,  
it is possible to recognize earth surface features and objects  
without any direct contact (10).  
Porphyry ore deposits can mainly indicate  
mineralization and alteration zonations (11, 12). Most of the  
copper porphyry minerals have a distribution pattern resulted  
from mineralization and alteration of deposits (6, 13). This  
distribution pattern of hydrothermal alteration zones can be  
used as an index to enhance and explore the copper porphyry  
minerals using RS techniques (14, 15, 7). In the recent years,  
it has been attempted to use spectral characteristics of  
satellite images in alteration zones and to employ different  
methods of image processing the alteration mapping (12).  
In this research, it has been attempted to use different  
techniques to discriminate alteration zones with regard to  
previous studies and lithologic characteristics of the region  
in order to show promising areas for exploration of porphyry  
Corresponding author: Mohammad Ali Nezammahalleh,  
PhD in geomorphology, University of Tehran, Tehran, Iran,  
Email: mnezammahalleh@ut.ac.ir, Tele: +989108350107.  
3
53  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
copper.  
Anar-Baft and Baft-Bam. This contains 95% of discovered  
reserves of copper in Iran. Deposit generation models  
discuss forming process of a deposit or a mineralization zone  
as well as examine activating, transporting, and  
concentrating forces on the elements. A set of the models as  
conceptual bases for exploration-geologic data may be  
simultaneously selected for analysis. Although main  
potential in the region of Anar-Baft metallogenic belt is  
porphyry copper, but it is also important for vein copper  
mineralization, poly-metal and vein lead and zinc. (Figure  
1).  
1
.2 Geographic and geologic position of Chahr-Gonbad  
The study area of this research is Chahr-Gonbad (from  
5
6° 28´ E to 56° 08´ E and from 29° 25´ N to 29° 40´ N),  
Kerman, Iran. In geologic map, the study area is located in  
south part of central Iran on Urmia-Dokhtar geologic zone,  
a metamorphic formation rich in cropped porphyry copper  
and iron. There are plenty of villages linked by a good  
transportation network in this mountainous region. Kerman  
metallogenic belt is south part of the magmatic-metallogenic  
unit of Urmia-Dokhtar zone and divided into two sections of  
Figure 1: location of the study area  
2
Materials and method  
2
.2 Spectral response of hydrothermal alterations and  
2
.1 Processing the region image  
detection of the minerals on etm+ and oli images  
The minerals of hydrothermal alteration zones can be  
detected on a special range of electromagnetic spectrum  
mainly in visible and near-Infrared. Because of wide  
bandwidth of the electromagnetic spectrum of ETM+ and  
OLI images, they are not able to discriminate separate  
minerals (16). Nevertheless, they can easily discriminate the  
minerals in hydrothermal alteration zones in near and middle  
infrared.  
We have used Landsat 7 and 8 images captured by  
Enhanced Thematic Mapper (ETM) sensor in eight bands  
and Operational Land Imager (OLI) in 7 bands for surveying  
and exploring hydrothermal alterations of porphyry minerals  
16, 17, 18). The study area is located in a scene with path  
60 and row 40 of the satellite image. The orthorectified  
ETM and OLI images of the study area for the date July 22,  
003, and June 1, 2019, with 0% of cloud cover has been  
taken from Landsat.org. Its geometric precision has also  
been controlled by topographic map. To eliminate  
atmospheric effects including water vapor and aerosols, we  
have used black targets like mountain shadow.  
(
1
2
The minerals characterizing the hydrothermal alteration  
are usually applied in exploration of different minerals,  
particularly porphyry, using satellite images. These minerals  
3
54  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
are grouped into three categories: hydroxyl (mica and clay),  
sulfate dihydrate (gypsum and alunite), and iron-bearing  
hydroxyl or ferrous minerals (Hematite, Geothite, Jarosite).  
The minerals mainly detectable in infrared range are  
belonging to hydroxyls and sulfates. The iron-oxide is often  
found in alteration zones and surface outcrops due to  
weathering. Thus, it can be important as a key to show the  
areas capable of mineralization. The existence of the iron  
oxide in rocks makes them red, brown, and orange in color.  
Existence of clay minerals gives light color to the rocks (19).  
Factor (OIF).  
2.4 Band ratios  
The ratios of bands are obtained by dividing DNs in one  
spectral band by those related to another band in each pixel.  
The new image is produced as the ratio of values of one band  
to those of another. One advantage of the ratio images is that  
they represent the color and spectral characteristics of earth  
surface features regardless of luminance effects due to  
various topographies. Indeed, they have no albedo effects of  
the primary data (20). Remote sensing techniques using the  
band ratio images have been used for enhancement of  
hydrothermal alteration zones as well as vegetation covers in  
many studies (14, 15, 13, 3).  
We have tested different band ratios in this study to  
enhance hydrothermally altered rocks and lithological units.  
The combinations of bands are selected based on spectral  
reflectance and position of the absorption bands of the  
minerals going to be mapped. The following band ratios are  
usually used for geological use to discriminate lithological  
features (16). The band ratios are 4/2 for iron oxide, 6/7 for  
hydroxyl bearing rock, and 7/5 for clay minerals. We have  
presented the laboratory spectral signatures of some minerals  
(Figure 3). Dependent on the spectral characteristics of the  
minerals, we can use band ratios of 3/1 and 5/7 in ETM data  
to detect iron-oxide and hydroxyl minerals, respectively.  
(Figure 2).  
Remote sensing techniques have been used for many  
years to produce the alteration mapping. Today, common  
image processing techniques are also applied for  
enhancement and detection of alteration zones. All the data  
and methods have been employed as supplementary to better  
find the promising areas in Zone 4, Chahr Gonbad, as the  
most important porphyry copper in Iran.  
2
.3 Image difference and band color composite  
Giving one of the three colors of red, green, and blue to  
a combination of three bands creates color composite image.  
There are many studies about the best selection of the colors  
to show maximum information of the minerals (16). The  
goal of selecting suitable combination of the bands to  
produce colored images is to minimize the nonsignificant  
information and maximize use of valuable information. One  
method of selecting the color composite is Optimized Index  
Figure 2: Spectral range of sensors ETM+, OLI, TIRS according to Landsat Website  
3
55  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
Figure 3: (A) Laboratory spectra of alunite, chlorite, kaolinite, muscovite, calcite, and epidote. (B) Laboratory spectra of limonite, jarosite,  
hematite and goethite (Beiranvand Pour; 2015).  
2
.5 Standard Principal Component Analysis (PCA) and  
Crosta  
In a look at different bands of a satellite image, e.g.,  
visible and middle-Infrared bands for the purpose.  
2.6 Matched filtering  
ETM+ and OLI, they seem very similar. This similarity is  
resulted from correlated information in the bands.  
Processing of the highly correlated bands is usually difficult  
in multi-spectral images. There is a great proportion of  
redundant information in different bands of one image.  
Rendering the bands uncorrelated helps us to compact the  
useful information in the multi-spectral images. Using the  
PCA, we can remove or reduce the redundant information  
and compress them into a coordinate system without any loss  
of information. In the analysis, the first principal component  
explains the highest variance of the data. The main  
application of the PCA is to reduce the variables called  
dimensionality reduction. (21)  
The Matched Filtering (MF) is a technique increasing  
reference member and minimizing reflectance of undefined  
background (22). However, all the methods for alteration  
mapping have  
a distinction function attempting to  
discriminate alteration minerals using a certain algorithm. In  
MF, we initially received the corrected bands and then as  
another input we integrated that with the spectral signature  
of the minerals going to be discriminated. We have  
compared the net spectrums of the minerals from spectral  
library with the spectrums of the same minerals of this study  
area. The highest spectral coincidence of the minerals with  
the library is considered as a recognized class.  
Using the PCA method, we can examine eigenvector  
loadings of multi-spectral images to discover the spectral  
signature of the minerals covered in an image. This is called  
Crosta technique as a simple way for alteration mapping by  
Landsat images. Combination of the PCA and Crosta has  
been applied by many studies for the detection of alterations  
in metallogenic belts (13). We have used 6 bands including  
2
Results and discussion  
The results of color composite of the bands 1, 3, 5 of  
Landsat ETM+ and OLI can give us a clear look of the  
alteration zones of the region. The Figure 4 shows color  
composite of the bands 1, 3, 5 of Landsat ETM+ and OLI. In  
this figure, the alteration zones are showed as white and  
yellow. Differential band composite of OLI sensor as (2-4)  
3
56  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
(
3-5) (7-6) by RGB can also well show the alteration zones  
variance of ETM+. The remarkable brightness of the region  
represents high correlation among the 6 bands. The PC2  
indicates that visible bands of the eigenvector have negative  
sign and Infrared bands have positive sign. In fact, the  
second PC is indicative of the differences between the visible  
and middle Infrared bands. Therefore, it is expected that  
materials with the highest spectral reflectance in visible  
spectrum are appeared in dark color and the materials with  
the highest spectral reflectance in middle Infrared spectrum  
are appeared in light color. In PC3 the highest eigenvector is  
related to the band 4. This indicates that in the PC vegetation  
is dominant feature of the surface.  
as illustrated in orange and pink color in Figure 5. The  
vegetation in red color is discriminated from alteration  
zones. The results of the band ratios are not clear for iron-  
oxide and not able to discriminate solely the hydroxyl  
minerals from vegetation. The band ratios of 3/1 and 4/2  
show iron-oxide areas and band ratios of 5/7 and 6/7 show  
hydroxyl areas in the region. In Figure 6, the alteration  
minerals are represented in white color pixels. For OLI data  
we can also use band ratio 4/2 for iron-oxide and 7/6 for  
hydroxyls. Results of the PCA in eigenvectors and  
eigenvalues of the images related to Chahr-Gonbad in 6  
bands of ETM+ have revealed that which of the spectral  
characteristics of the rocks, vegetation, and soil can account  
for the statistical variance of the PCs in the region (Table 1).  
The first PC with a positive weight contains 91% of the  
Figure 4: Color composite of the bands 1, 3, 5 of ETM+  
3
57  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
Figure 5: Differential band composite (2-4)(3-5)(7-6) of OLI  
The spectral information of clay and iron-oxide in  
principal components of ETM+ are mainly concentrated in  
PC5 and PC4. The PC4 with high loadings and opposite sign  
is discerned in bands 5 and 7. In the band 7 the hydroxyls  
have high absorption with loadings of -0.668 and in band 5  
the minerals have high reflectance with loadings of 0.625.  
Thus, in PC4 the hydroxyl zones can be seen in light color  
pixels.  
The results of the PCA method in 6 bands of visible and  
non-thermal infrared of ETM+ indicate that hydroxyls and  
iron-oxide can be mapped in PC4 and PC5, respectively. In  
PC5 band 1, in which iron-oxide shows higher absorption,  
has positive loadings of 0.612 and band 3, in which the  
mineral shows higher reflectance, has negative reflectance of  
area  
Eigenvector Band 1 Band 2 Band 3 Band 4 Band 5 Band 7  
pc 1  
pc 2  
pc 3  
pc 4  
pc 5  
pc 6  
0.191219 0.284945 0.446248 0.375444 0.559546 0.478597  
0.458172 0.408835 0.502328 0.034354 -0.52169 -0.31187  
0.122145 0.102285 0.158869 -0.89479 0.053124 0.38199  
0.188382 0.250972 -0.12651 -0.21635 0.625795 -0.66865  
0.612015 0.243922 -0.68638 0.10204 -0.09056 0.276067  
0.573188 -0.78658 0.190193 0.000119 0.110042 -0.06679  
We have used PC5 to show the iron-oxide as light color  
pixels. (Figure 7). We have combined the three images of  
principal components by RGB to produce a colored image.  
We have used PC4 for hydroxyl mapping, PC4+PC5 for  
hydrothermal alteration mapping, and PC5 for iron-oxide  
-
0.686. Thus, the iron-oxide zones are appeared in dark  
color pixels in PC5.  
Table 1: Results of PCA in 6 bands of ETM+ of the study  
3
58  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
mapping. The alteration areas have been displayed as white,  
light blue, and yellow (Figure 8).  
Figure 6: Band ratios 3/1 and 4/2 as well as 5/7 and 6/7 for detection of iron-oxide and hydroxyl  
Figure 7: Hydroxyl minerals (light color pixels) in PC4 and iron-oxide minerals (light color pixels) in PC5  
s
3
59  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
Figure 8: Colored RGB image (PC4, PC4+ PC5, PC5), alteration zones can be seen in white, light blue and yellow  
Figure 9: PC4 for hydroxyls (dark pixels and PC5 for iron-oxide (light pixels)  
3
60  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
Figure 10: PC4 for iron-oxide minerals (light pixels) and PC5 for hydroxyl minerals (dark color)  
Table 2 presents the results of PCA for 7 bands of OLI.  
PC4 show the highest value for clay minerals in bands 7 and  
appropriate bands of high information, for example, near  
infrared (VNIR) for iron-oxide and shortwave infrared  
(SWIR) for argillite and propylitic zone. Reducing the  
number of bands in PCA ensure that some features are  
excluded to better find some special features of interest. This  
technique called Crosta makes it more likely to discover the  
minerals of the study. We have conducted two PCAs in this  
research. In one analysis, we included bands 1, 3, 4, and 5  
for extraction of iron-oxide and excluded band 7 to prevent  
hydroxyl mapping. In the other analysis, we included bands  
of 1, 4, 5, and 7 for extraction hydroxyl minerals and  
excluded bands 2 and 3 to prevent iron-oxide mapping  
6
. Thus, we can see the clay minerals in dark color in this  
PC4. In PC5, the iron-oxide in bands 2 and 4 has the highest  
difference and appeared in light color. (Figure 9)  
Table 2: Principal Component Analysis on 6 bands of OLI  
in the study area  
Band 1 Band 2 Band 3 Band 4 Band 5 Band 6 Band 7  
(Table 3). All statistical information including eigenvector,  
correlation matrix, and eigenvalues have been analysed for  
both the groups of bands. Given the higher spectral  
reflectance of iron-oxide in band 3 and higher absorption in  
band 1 along with the coefficients of eigenvectors, the areas  
spotted by the iron-oxide are discriminated by light color on  
PC4 (0.673 and -0.722) (Figure 10 and 11). The higher  
spectral reflectance of hydroxyl minerals in band 4 and  
higher absorption in band 7 along with the coefficients of the  
eigenvectors are discriminated the areas spotted by hydroxyl  
as dark color on the PC4 (0.670 and -0.609) (Figure 12).  
pc 1  
pc 2  
pc 3  
pc 4  
pc 5  
pc 6  
(Table 4).  
Table 3: eigenvector of PCA for iron-oxide  
Eigenvector Band 1 Band 3 Band 4 Band 5  
PC 1  
PC 2  
PC 3  
-0.21433 -0.43232 -0.56632 -0.66816  
-0.48179 -0.4385 -0.30336 0.695393  
0.447459 0.408431 -0.75974 0.236137  
PC 4  
-0.7223 0.673795 -0.10026 -0.11929  
Table 4: eigenvector of PCA for hydroxyls  
Eigenvector Band 1  
Band 4  
Band 5  
Band 7  
Band 1  
Band 3  
Band 5  
Band 6  
-0.1875 -0.50969 -0.62884 -0.55644  
-0.28778 -0.78853 0.389394 0.379186  
0.854177 -0.30267 -0.28541 0.311954  
-0.39039 0.163824 -0.60949 0.670284  
pc 7  
In the Crosta selective PCA method, we have applied the  
3
61  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
As reference data to recognize iron-oxides and clay  
minerals, we have compared spectral signature of 7 minerals  
including Hematite, Geothite, Calcite, Illite, Jarosite,  
Kaolinite, and Limonite. The references have been received  
from USGS spectral collection. To compare the existing  
spectral bands from USGS with OLI and ETM images, the  
laboratory spectrums need to be resampled for every band of  
the image. The results of the previous methods show good  
consistence with this method. (Figure 11).  
mostly concentrated in central, west, and southeast portions  
of the study area in a southeast-northwest orientation.  
According to these results, the iron-oxide is also  
concentrated mainly in southeast and central parts of the  
study area. Indeed, the exploration results, whether exact,  
cannot substitute direct observation of geologic features.  
However, there may be no mineralization in the detected  
areas or they may not be so economical. Therefore, for  
evaluation of the resulted areas and discriminating the false  
anomalies from the real, it is necessary to make a field survey  
control of the anomalies.  
2
Conclusions  
Formation of mine reserves as a complicated geologic  
process is influenced by a variety of environmental factors.  
Many geologists and mine specialists classified various  
processes of development of minerals. Hence, the evidence  
and signs of existence of minerals are well recognized for  
exploration and mining activities. Given that the alteration  
mapping can be used as an indicator of exploration,  
discovering the alteration zones can be very useful for  
exploration activities in regional level. Using band  
composite, band ratio, PCA, Crosta and matched filtered  
methods, we have processed ETM+ and OLI images of  
Chahr-Gonbad and detected alteration minerals in relation  
with mineralization of porphyry copper. This is greatly  
consistent with the findings of Ercan et al. (2016) about the  
relation of alteration minerals with the mineralization. The  
results of the analyses have documented that the alterations  
are mainly argillite and propylitic and that the hydroxyls are  
Figure 11: spectral response of the minerals of the study  
using MF  
Figure 12: Mapping of iron-oxide areas with red color and clay minerals with green color  
3
62  
Journal of Environmental Treatment Techniques  
2020, Volume 8, Issue 1, Pages: 353-363  
1
1
1
0. Fujisada, H., Iwasaki, A., HARA, S., (2001) ASTER stereo  
system performance, proceeding of SPIE, the international  
society for optical engineering, 4540:39-40  
1. Abrams, M.J., Brown, D., Lepley, L., Sadowski, R., (1983),  
Remote sensing for porphyry copper deposits in Southern  
Arizona, Economic Geology 78, 591-604  
2. Ruiz Armenta, J.R. and Prol-Ledesma, R.M., (1998);  
Techniques for enhancing the spectral response of  
hydrothermal alteration minerals in Thematic Mapper images  
of Central Mexico, International Journal of Remote Sensing  
Aknowledgment  
The authors are grateful to Hassan Nezammahalleh for  
providing useful advices for English editions.  
Ethical issue  
Authors are aware of, and comply with, best practice in  
publication ethics specifically with regard to authorship  
(avoidance of guest authorship), dual submission,  
manipulation of figures, competing interests and compliance  
with policies on research ethics. Authors adhere to  
publication requirements that submitted work is original and  
has not been published elsewhere in any language.  
1
9, 1981-2000.  
13. Ercan, Hatice Ünal ;  
Ü. Işik Ece;  
Paul A. Schroeder;  
Zekiye Karacik, (2016) Differentiating Styles of Alteration  
within Kaolin-Alunite Hydrothermal Deposits of Çanakkale,  
NW Turkey, Clays and Clay Minerals 64, 245-274.  
1
4. Rowan, L.C., Goetz, A.F.H. and Ashley, R.P., (1979)  
Discrimination of hydrothermally altered and unaltered rocks  
in the visible and near infrared: geophysics 42, 533-535.  
Competing interests  
The authors declare that there is no conflict of interest  
that would prejudice the impartiality of this scientific work.  
15. Podwysocki, M.H., Segal, D.B. and Abrams, M.J., (1983) Use  
of multispectral scanner images for assessment of  
hydrothermal alteration in the Marysvale, Utah mining area,  
Economic Geology 78, 675-687.  
Authors’ contribution  
All authors of this study have a complete contribution  
for data collection, data analyses and manuscript writing.  
1
1
1
1
6. Beiranvand Pour Amin; Mazlan Hashim; (2015) Hydrothermal  
alteration mapping from Landsat-8 data, Sar Cheshmeh copper  
mining district, south-eastern Islamic Republic of Iran; Journal  
of Taibah University for Science 9; 155-166.  
7. Bennet, S.A., Atkinson, W.W. and Kruse, F.A., (1993) Use of  
Thematic Mapper imagery to identify mineralization in the  
Santa Tersa district, Sonora, Mexico: International Geology  
Review 35, 1009-1029.  
8. Chavez, P. S., Kwarteng, A. Y., (1989) extracting spectral  
contrast in Landsat Thematic Mapper image data using  
selective principal component analysis. Photogrammetric  
Engineering and Remote Sensing 53, 339-348.  
9. Tangestani, M.H. and Moore, F., (2000); Iron-oxide and  
hydroxyl enhancement using the Crosta method: A case study  
from the Zagros Belt, Fars province, Iran, Journal of Applied  
Geosciences 2, 140 -146.  
0. Sabins Rowan, F.F., (1996) Remote Sensing (Principle and  
Interpretation), W.H.Freeman Co., New York, PP 575  
1. Crosta, A.P. and Moore, McM., (1989); Enhancement of  
Landsat Thematic Mapper imagery for residual soil mapping  
in SW Minais Gerais State, Brazil: A prospecting case history  
in Greenstone belt terrain, Proceeding of the 7th ERIM  
thematic conference: Remote Sensing For Exploration  
Geology, 1173-1187.  
2. Harsanyi, Joseph C.; Chang, Chein-I; (1994) Hyperspectral  
Image Classification and Dimensionality Reduction: An  
Orthogonal Subspace Projection Approach, IEEE  
Transactions on Geoscience and Remote Sensing 32(4), 779 –  
References  
1
.
Jiang, Chen; Mahyar Yousefi; Yan Zhao; Chunpeng Zhang; Sen  
Zhang; Zhaoxia Mao; Mingshen Peng; Renping Han; (2019)  
Modelling ore-forming processes through a cosine similarity  
measure: Improved targeting of porphyry copper deposits in  
the Manzhouli belt, China; Ore Geology Reviews 107, 108-  
1
18.  
2
3
.
.
Tursunai Bektemirova; Apas Bakirov; Ruizhong Hu; Hongping  
He; Yuanfeng Cai; Wei Tan; Aiqing Chen; (2018)  
Mineralogical evolution of the paleogene formations in the  
Kyzyltokoy basin, Kyrgyzstan: implications for the formation  
of glauconite, Clays and Clay Minerals 66 (1): 43-60  
Ferrier Graham; Athanassios Ganas; Richard Pope, (2019)  
Prospectivity mapping for high sulfidation epithermal  
porphyry deposits using an integrated compositional and  
topographic remote sensing dataset; Ore Geology Reviews  
2
2
1
07, 353-363.  
4
5
.
.
Curan, P.J. (1986) Principle of remote sensing; Longman  
Scientific & Technical, New York, 282.  
Percival, Jeanne B.; Sean A. Bosman; Eric G. Potter; Jan M.  
Peter; Alexandra B. Laudadio; Ashley C. Abraham; Daniel A.  
Shiley; Chris Sherry; (2018) Customized Spectral Libraries for  
Effective Mineral Exploration: Mining National Mineral  
Collections, Clays and Clay Minerals 66, 297-314.  
2
6
7
.
.
(6) Salehi, Touba; Majid H. Tangestani (2018), Large-scale  
mapping of iron oxide and hydroxide minerals of Zefreh  
porphyry copper deposit, using Worldview-3 VNIR data in the  
Northeastern Isfahan, Iran; International Journal of Applied  
Earth Observation and Geoinformation 73, 156-169.  
Tapia Joseline; Brian Townley; Loreto Córdova; Fernando  
Poblete; César Arriagada; (2016) Hydrothermal alteration and  
its effects on the magnetic properties of Los Pelambres, a large  
multistage porphyry copper deposit; Journal of Applied  
Geophysics 132, 125-136.  
7
85  
8
9
.
.
Rowan, L.C., Mars, J.C., (2003) Lithologic mapping in the  
Mountain Pass, California area using Advanced Spaceborne  
Emission and Reflection Radiometer (ASTER) data, Remote  
Sensing of Environment 84, 350–366.  
Kaufmann, H.J., (1988) Mineral exploration along the Aqaba-  
Levant structure by use of TM-data-concepts, processing and  
results, International Journal of Remote Sensing 9, 1639-1658  
3
63