See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/258201119

Identification of hydrothermal alteration minerals for exploring of porphyry

copper deposit using ASTER data, SE Iran

Article  in  Journal of Asian Earth Sciences · November 2011

DOI: 10.1016/j.jseaes.2011.07.017

CITATIONS READS

78 870

2 authors:

Amin Beiranvand Pour Mazlan Hashim

Korea Polar Research Institute Universiti Teknologi Malaysia
122 PUBLICATIONS   1,022 CITATIONS    341 PUBLICATIONS   1,926 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Satellite-based water yield mapping View project

sub_pixel unmixing View project

All content following this page was uploaded by Amin Beiranvand Pour on 19 October 2017.

The user has requested enhancement of the downloaded file.

 Journal of Asian Earth Sciences 42 (2011) 1309–1323

Contents lists available at SciVerse ScienceDirect

Journal of Asian Earth Sciences

journal homepage: www.elsevier.com/locate/jseaes

Identification of hydrothermal alteration minerals for exploring of porphyry

copper deposit using ASTER data, SE Iran
Amin Beiranvnd Pour, Mazlan Hashim ⇑
Institute of Geospatial Science & Technology (INSTeG), Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor Bahru, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: The NW–SE trending Central Iranian Volcanic Belt hosts many well-known porphyry copper deposits in
Received 1 June 2010 Iran. It becomes an interesting area for remote sensing investigations to explore the new prospects of por-
Received in revised form 5 July 2011 phyry copper and vein type epithermal gold mineralization. Two copper mining districts in southeastern
Accepted 13 July 2011
segment of the volcanic belt, including Meiduk and Sarcheshmeh have been selected in the present study.
Available online 6 August 2011
The performance of Principal Component Analysis, band ratio and Minimum Noise Fraction transforma-
tion has been evaluated for the visible and near infrared (VNIR) and, shortwave infrared (SWIR) subsys-
Keywords:
tems of ASTER data. The image processing techniques indicated the distribution of iron oxides and
ASTER
Copper exploration
vegetation in the VNIR subsystem. Hydrothermal alteration mineral zones associated with porphyry cop-
Alteration zones per mineralization identified and discriminated based on distinctive shortwave infrared (SWIR) proper-
Principal Component Analysis ties of the ASTER data in a regional scale. These techniques identified new prospects of porphyry copper
Band ratio mineralization in the study areas. The spatial distribution of hydrothermal alteration zones has been ver-
Minimum Noise Fraction ified by in situ inspection, X-ray diffraction (XRD) analysis, and spectral reflectance measurements.
Results indicated that the integration of the image processing techniques has a great ability to obtain sig-
nificant and comprehensive information for the reconnaissance stages of porphyry copper exploration in
a regional scale. The results of this research can assist exploration geologists to find new prospects of por-
phyry copper and gold deposits in the other virgin regions before costly detailed ground investigations.
Consequently, the introduced image processing techniques can create an optimum idea about possible
location of the new prospects.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction System AM-1 (EOS AM-1) polar orbiting spacecraft in December

1999. The ASTER instrument is a cooperative effort between the
Mapping surface mineralogy using remote sensing sensors pro- Japanese Ministry of Economic Trade and Industry (METI) and Na-
vides an opportunity to improve initial steps of ore deposits explo- tional Aeronautics and Space Administration (NASA). ASTER is a
ration (Sabins, 1999; Rowan et al., 2003; Mars and Rowan, 2006; Di multispectral imaging Sensor that measures reflected and emitted
Tommaso and Rubinstein, 2007; Zhang et al., 2007; Moghtaderi electromagnetic radiation from Earth’s surface and atmosphere in
et al., 2007; Gabr et al., 2010). Hydrothermal fluid processes that 14 bands (Yamaguchi et al., 1999; Abrams, 2000). There are three
alter the mineralogy and chemical composition of the country groups of bands: (i) three recording visible and near infrared
rocks generate porphyry ore deposits. The altered rocks having (VNIR) ranging between 0.52 and 0.86 lm at a spatial resolution
diagnostic spectral absorption features due to produced mineral of 15 m; (ii) six recording portions of shortwave infrared (SWIR)
assemblages (Hunt and Ashley, 1979). Exploration geologists have from 1.6 to 2.43 lm at a spatial resolution of 30 m; and (iii) five
used sophisticated remote sensors to detect hydrothermal alter- recording thermal infrared (TIR) in the 8.125–11.65 lm wave-
ation mineral assemblages for reconnaissance stages of porphyry length region with resolution of 90 m. An additional backward-
copper and gold exploration (Rowan et al., 2006; Gersman et al., looking band in the VNIR makes it possible to construct digital
2008; Bedini et al., 2009; Gabr et al., 2010; Pour et al., 2011). The elevation models from bands 3 and 3b. ASTER swath width is
Advanced Spaceborne Thermal Emission and Reflection Radiome- 60 km (each scene is 60  60 km) and useful for regional mapping
ter (ASTER) sensor presents unprecedented opportunities for min- (Fujisada, 1995). ASTER sensor allows the discrimination and iden-
eral exploration. It was launched on NASA’s Earth Observing tification of hydrothermal alteration minerals in the SWIR region
(Abrams and Hook, 1995). The ASTER bands in the SWIR were spe-
⇑ Corresponding author. Tel.: +60 7 5530666; fax: +60 7 5531174. cifically selected to highlight the presence of spectral absorption
E-mail addresses: beiranvand.amin80@gmail.com, mazlanhashim@utm.my,
features present in minerals, such as clay, carbonates, sulfates
profmhashim@gmail.com (M. Hashim). and other hydrous phases due to overtones and combination tones

1367-9120/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jseaes.2011.07.017
1310 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

of fundamental absorptions of Al–O–H, Mg–O–H, Si–O–H, and CO3

(Hunt, 1977; Abrams, 2000). Moreover, the ASTER visible and near
infrared and thermal infrared data can provide adequate capability
for remote identification of vegetation and iron oxide minerals in
surface soil, and mapping carbonates and silicates, respectively
(Bedell, 2001; Ninomiya, 2003a,b; Rockwell and Hofstra, 2008).
Ideal porphyry copper deposits are typically characterized by
hydrothermal alteration mineral zones. The core of quartz and
potassium-bearing minerals is surrounded by multiple alteration
zones (Fig. 1; Lowell and Guilbert, 1970). The broad phyllic zone
is characterized by illite/muscovite (sericite) that indicates an in-
tense Al-OH absorption feature centered at 2.20 lm, coinciding
with ASTER band 6. The narrower argillic zone including, kaolinite
and alunite displays a secondary Al-OH absorption feature at
2.17 lm that corresponds with ASTER band 5. The mineral assem-
blages of the outer propylitic zone are epidote, chlorite, and
calcite that exhibit absorption features situated in the 2.35 lm,
which coincide with ASTER band 8 (Fig. 2) (Mars and Rowan,
2006).
The objective of this paper is to utilize ASTER data for mapping
hydrothermal alteration mineral zones associated with porphyry
copper mineralization in two large mining districts, including
Sarcheshmeh and Meiduk in the southeastern segment of the
NW–SE trending Central Iranian Volcanic Belt, southeast Iran. It
becomes an interesting area for remote sensing investigations to
explore the new prospects of porphyry copper and vein type
epithermal gold mineralization. The major part of this belt has
well-exposed and sparse vegetated surface, which is ideal for
remote sensing investigations. In this belt, the abundance of
known porphyry copper deposits reflects the potential economic
importance and warrants the exploration of the new prospects.
This paper emphasizes on the specific robust, fast, simple and
highly-effective spectral transform techniques for the identifica-
tion of hydrothermal alteration minerals. The selected transform
methods are Principal Component Analysis (PCA), specialized band
ratio (BR) and Minimum Noise Fraction (MNF). The following study Fig. 2. Laboratory spectra of muscovite, kaolinite, alunite, epidote, calcite, and
hypotheses are then set: (i) hydrothermal alteration mineral zones chlorite re-sampled to ASTER bandpasses. Spectra include muscovite, typical in
associated with porphyry copper mineralization can be detected phyllic alteration zone, with a 2.20 lm absorption feature; kaolinite and alunite,
which are common in argillic alteration zone, have 2.17 lm absorption features;
using ASTER VNIR and SWIR bands; and (ii) three spectral trans-
and epidote, calcite, and chlorite, which are typically associated with propylitic
form techniques, namely PCA, BR and MNF could adequately detect alteration zone and display 2.35 lm absorption features (Clark et al., 1993b; Mars
hydrothermal alteration zones associated with porphyry copper and Rowan, 2006).
deposit using exclusively ASTER VNIR and SWIR data in a regional
scale.

Fig. 1. Hydrothermal alteration zones associated with porphyry copper deposit (modified from Lowell and Guilbert, 1970; Mars and Rowan, 2006). (A) Schematic cross
section of hydrothermal alteration mineral zones, which consist of propylitic, phyllic, argillic, and potassic alteration zones. (B) Schematic cross section of ores associated with
each alteration zone.
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1311

2. Materials and methods deposit is within a belt of Eocene volcanic rocks and Oligo-Miocene
subvolcanic granitoid rocks. The oldest host rocks at the
2.1. Geological setting Sarcheshmeh porphyry copper deposit belong to an Eocene volca-
nogenic complex, also known as the Sarcheshmeh complex. The
Iran is a semi-arid country located on the Tethyan Copper Belt. complex consists of pyroxene trachybasalt, pyroxene trachyande-
It has great potential for exploration porphyry copper and gold site of potassic and shoshonitic affinity (Aftabi and Atapour,
deposits using remote sensing instruments. Fig. 3 shows NASA’s 1997), less abundant andesite and rare occurrences of agglomerate,
2000 GeoCover global orthorectified Landsat 7 mosaics of Iran as tuff, and tuffaceous sandstone. These were intruded by a complex
color composite of band 7 as red, band 4 as green, and band 2 as series of Oligo-Miocene granitoid intrusive phases such as quartz
blue (Merged MrSID files from http://zulu.ssc.nasa.gov/mrsid). diorite, quartz monzonite and granodiorite. The granitoid rocks
Green color is as indicator of vegetated surface. Many of known are cut by a series of intramineral hornblende porphyry, feldspar
porphyry copper mineralization occurs in the Central Iranian Vol- porphyry and biotite porphyry dykes (Waterman and Hamilton,
canic Belt (Hassan-Nezhad and Moore, 2006). This Belt (trends 1975). Hydrothermal alteration and mineralization at Sar-
NW–SE) is the most important volcano-plutonic complex with tre- cheshmeh are centered on the stock and were broadly synchronous
mendous economic potential for copper mineralization which with its emplacement. Early hydrothermal alteration was domi-
formed by subduction of the Arabian plate beneath central Iran nantly potassic and propylitic, and was followed later by phyllic,
during the Alpine orogeny (Berberian and King, 1981; Shahabpour, silicic and argillic alteration (Hezarkhani, 2006). The Meiduk por-
2005, 2007). This study focuses on southeastern segment of the phyry copper deposit (55°100 0500 E, 30°250 1000 N) is located 45 km
NW–SE trending Central Iranian Volcanic Belt. The Study area is lo- northeast of Shahr-e-Babak, Kerman province. Fig. 6 shows geolog-
cated in black cubic in southeastern part of the Fig. 3. In this area, ical map of the Meiduk area. Lower Eocene rocks are parts of a vol-
yearly precipitation average is 25 cm or less (Modarres and Da canic complex consisting of rhyolite lavas, breccias, acidic tuffs,
Silva, 2007), thus earth’s surface has well-exposed due to very and pyroclastic rocks. This volcanic complex followed by another
sparse to nonexistent vegetation cover, which is quite suitable Eocene to Oligocene volcanic complex, which consists of trachyba-
for remote sensing studies. Simplified geologic map of the south- salt, andesite and trachyandesite, andesite–basalt, and acidic tuff.
eastern segment of the Central Iranian Volcanic Belt illustrates The intrusive rocks were emplaced into the volcanic rocks as stock
the location of study areas, rock units, and general structural dykes at the Meiduk ore deposit. The volcanic complexes and
geology (Fig. 4). The study areas include Sarcheshmeh and Meiduk intrusive rocks are partly covered by late Miocene–Pliocene volca-
porphyry copper deposits located in the southern part of the nic and subvolcanic rocks of the Masahim stratovolcano. The youn-
volcano-plutonic belt where Cu and Mo are currently mined. The gest volcanic rocks in the study area are Quaternary in age and
Sarcheshmeh porphyry copper deposit (55°520 2000 E, 29°580 4000 N) range from trachyte to dacite (Hassanzadeh, 1993). The Cu-miner-
is the most important copper ore deposit in Iran, 160 km south- alization and associated hydrothermal alteration zones are focused
west of Kerman city and is one of the largest porphyry copper on the Miocene dioritic Meiduk porphyry and Eocene andesitic
deposits in the world. Fig. 5 shows geological map of the rocks (Boomeri et al., 2009). However, distribution of the various
Sarcheshmeh area (Hubner, 1969a; Mars and Rowan, 2006). The alteration types is irregular. The concentric alteration zones from
the center outward are potassic, phyllic, and propylitic. This pat-
tern is similar to alteration envelopes that are associated with
many other porphyry Cu–Mo deposits. Amraie (1991) recognized
potassic, phyllic, argillic, and propylitic alteration zones at Meiduk
area.

2.2. Preprocessing of ASTER data

The ASTER data used in this study were obtained from the Earth
and Remote Sensing Data Analysis Center (ERSDAC) Japan, and
consist of two cloud-free level 1B ASTER scenes of the study sites
in southeastern part of the Central Iranian Volcanic Belt. They were
acquired on July 15, 2007 for the Sarcheshmeh area, and June 20,
2006 for the Meiduk area, respectively. The level 1B data product
measures radiance at the sensor without atmospheric corrections,
and were produced from the original level 1A format by ERSDAC
(Abrams, 2000). The images were pre-georeferenced to UTM zone
40 North projection with using the WGS-84 datum. The crosstalk
correction performed to both data sets in this study, aimed at
removing the effects of energy overspill from band 4 into bands
5 and 9 (Hewson et al., 2005). We have performed this correction
by Cross-Talk correction software that available from
www.gds.aster.ersdac.or.jp. In addition, the ASTER Level 1B data
were converted to reflectance using the Internal Average Relative
Reflection (IARR) method (Ben-Dor et al., 1995). Ben-Dor et al.
(1995) recommended IARR reflectance technique for mineralogical
mapping as a preferred calibration technique; it does not require
the prior knowledge of samples that collected from the field. The
Fig. 3. NASA’s 2000 GeoCover global orthorectified Landsat 7 mosaics merged 30 m resolution SWIR and of the ASTER data were re-sampled to
image of Iran as color composite of band 7 as red, band 4 as green, and band 2 as
blue. The Study area is located in black cubic in southeastern part of image. (For
correspond to the VNIR 15-m spatial dimensions. Pan-sharpening
interpretation of the references to color in this figure legend, the reader is referred technique applied to enhance the spatial resolution because it does
to the web version of this article.) not effect on the pixel digital numbers due to re-sampling affects.
1312 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Fig. 4. Simplified geology map of southeastern segment of the Central Iranian Volcanic Belt (modified from Shafiei, 2010). Study areas are located in ellipsoidal polygons.

Fig. 5. Geological map of Sarcheshmeh region (modified from Hubner, 1969b; Mars and Rowan, 2006).

The ASTER surface reflectance scenes were processed and analyzed Principal Component Analysis (PCA), band ratio (BR), and Mini-
by ENVI (Environment for Visualizing Images) version 4.5 and mum Noise Fraction (MNF) performed over two full ASTER scenes.
ERMapper version 6.4 software packages.
2.3.1. Principal Component Analysis
2.3. Image processing techniques Principal Component Analysis (PCA) is a multivariate statistical
technique that selects uncorrelated linear combinations (eigenvec-
We have implemented robust, fast and reliable techniques tor loadings) of variables in such a way that each component suc-
based on spectral characteristics of alteration key minerals for a cessively extracted linear combination and has a smaller variance
systematic selective extraction of the information of interest. (Singh and Harrison, 1985; Chang et al., 2006). PCA is a well-known
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1313

Fig. 6. Geological map of Meiduk region (modified from Hubner, 1969a; Mars and Rowan, 2006).

method for lithological and alteration mapping in metalogenic where b4, b5, b6, b7, b8 and b9 designated for ASTER bands 5, 6, 7, 8
provinces (Crosta et al., 2003; Rowan and Mars, 2003; Ranjbar and 9, respectively.
et al., 2004; Kargi, 2007; Massironi et al., 2008; Moore et al.,
2008; Tangestani et al., 2008; Amer et al., 2010). Standard PCA 2.3.3. Minimum Noise Fraction
transformation has applied to the two full ASTER scenes of the Sar- The MNF transformation is used to determine the inherent
cheshmeh and Meiduk regions. A total of nine new image compo- dimensionality of image data, segregate noise in the data, and re-
nents are generated from the original nine bands (VNIR + SWIR) duce the computational requirements for subsequent processing
ASTER data. (Green et al., 1988; Boardman et al., 1995). The Minimum Noise
Fraction (MNF) transformation involves two steps; first, which also
2.3.2. Band ratio called noise-whitening, principal components for noise covariance
Band ratio (BR) is a technique where the DN value of one band is matrix are calculated; this step decorrelates and rescales the noise
divided by the DN value of another band. BRs are very useful for in the data. In the second step, the principal components are de-
highlighting certain features or materials that cannot be seen in rived from the noise whitened data. The data can then be divided
the raw bands (Inzana et al., 2003). ASTER BR technique has also into two parts: one part associated with large eigenvalues and
reported wide acceptance in geological mapping in the recent the other part with near unity eigenvalues and noise dominated
years (Gad and Kusky, 2007; Khan and Mahmood, 2008; Massironi images. Using data with large eigenvalues separates the noise from
et al., 2008; Amer et al., 2010; Aboelkhair et al., 2010). Selected AS- the data, and improves spectral results (Green et al., 1988). MNF
TER VNIR and SWIR bands have used for BR in this study. Three BRs analysis can identify the locations of spectral signature anomalies.
have performed (i) stabilized vegetation index (StVI) = (band 3/ This process is of interest to exploration geologist because spectral
band 2)  (band 1/band 2) for detecting vegetation features anomalies are often indicative of alterations due to hydrothermal
(Ninomiya, 2003a); (ii) ratio of band 4/2 for identifying iron oxides mineralization. MNF has applied on ASTER SWIR bands to discrim-
(Abdelsalam and Stern, 2000), (iii) ratio of band 7/6 for identifying inate hydrothermally altered rocks from surrounding igneous
muscovite (Hewson et al., 2005). The strategy of the above ratios is background in the study areas.
to identify vegetation, iron oxides, and OH-bearing mineral zones,
thus lead to an overall large area mapping of the hydrothermal
alteration areas associated with porphyry copper deposits. Relative 3. Results, analysis and discussion
Absorption Band Depth (RBD) is a useful three-point ratio formula-
tion for displaying Al–O–H, Fe, Mg–O–H, and CO3 absorption 3.1. Principal component analysis
intensities. For each absorption feature, the numerator is the sum
of the bands representing the shoulders, and the denominator is PCA outputs are presented as tables of statistic factors and se-
the band located nearest the absorption feature minimum (Crow- lected PC images from these transformations are reproduced in fig-
ley et al., 1989). Three RBD ratios have adopted in this study; ures to support the discussion. The image eigenvectors and
RBD5, RBD6, and RBD8 were assigned for RGB (red, green, and blue) eigenvalues obtained from PCA, using covariance matrix on all nine
color combination to delineate argillic, phyllic and propylitic reflective bands of ASTER of the Meiduk and Sarcheshmeh scenes
hydrothermal alteration zones. The RBD ratios have been derived are indicated in Tables 1 and 2.
based on Crowley et al. (1989) as follows: PC1 is composed of a positive weighting of all nine (VNIR + S-
WIR) total bands. As indicated by the eigenvalues, PC1 accounts
ðb4 þ b6Þ for 95.71% and 94.06% of the total variance for the data for Meiduk
RBD5 ¼ ð1Þ and Sarcheshmeh scene, respectively. Overall scene brightness, or
b5
albedo, is responsible for the strong correlation between multi-
ðb5 þ b7Þ spectral image bands (Loughlin, 1991). PCA has effectively mapped
RBD6 ¼ ð2Þ
b6 albedo into PC1 of the transformation. All of eigenvector loadings
for PC2 are negative (Tables 1 and 2). It is evident that PC2 is noisy
ðb7 þ b9Þ without any information. Eigenvector loadings for PC3 indicate
RBD8 ¼ ð3Þ
b8 that PC3 probably describes the difference between the visible
1314 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Table 1
Eigenvector matrix of principal components analysis on VNIR + SWIR bands of ASTER data for Meiduk scene.

Input bands Band 1 Band 2 Band 3 Band 4 Band 5 Band 6 Band 7 Band 8 Band 9 Eigen values (%)
PC1 0.320 0.322 0.320 0.325 0.325 0.325 0.322 0.323 0.323 95.71
PC2 0.004 0.012 0.003 0.120 0.091 0.067 0.153 0.142 0.122 2.72
PC3 0.430 0.422 0.414 0.164 0.024 0.123 0.436 0.360 0.273 0.89
PC4 0.100 0.124 0.417 0.421 0.224 0.456 0.430 0.088 0.414 0.31
PC5 0.289 0.249 0.573 0.245 0.037 0.034 0.439 0.276 0.437 0.14
PC6 0.270 0.269 0.386 0.519 0.117 0.486 0.204 0.331 0.180 0.12
PC7 0.044 0.111 0.087 0.316 0.170 0.087 0.254 0.638 0.607 0.07
PC8 0.019 0.126 0.189 0.299 0.721 0.484 0.072 0.247 0.190 0.03
PC9 0.698 0.698 0.022 0.069 0.088 0.071 0.001 0.037 0.064 0.01

Table 2
Eigenvector matrix of principal components analysis on VNIR + SWIR bands of ASTER data for Sarcheshmeh scene.

Input bands Band 1 Band 2 Band 3 Band 4 Band 5 Band 6 Band 7 Band 8 Band 9 Eigen values (%)
PC1 0.320 0.320 0.320 0.329 0.328 0.328 0.324 0.326 0.326 94.06
PC2 0.068 0.061 0.058 0.140 0.111 0.087 0.175 0.168 0.149 3.56
PC3 0.474 0.472 0.434 0.186 0.108 0.022 0.307 0.285 0.243 1.53
PC4 0.135 0.121 0.391 0.588 0.144 0.524 0.292 0.050 0.289 0.45
PC5 0.296 0.225 0.538 0.124 0.047 0.144 0.469 0.174 0.531 0.20
PC6 0.248 0.219 0.454 0.436 0.246 0.358 0.306 0.314 0.334 0.10
PC7 0.001 0.193 0.232 0.433 0.710 0.411 0.080 0.131 0.154 0.07
PC8 0.071 0.065 0.014 0.070 0.249 0.159 0.330 0.701 0.544 0.02
PC9 0.688 0.703 0.009 0.074 0.078 0.056 0.031 0.083 0.091 0.01

channels, including bands 1, 2, and 3 (negative eigenvector load- the PC5 image. Vegetation pixels follow the drainage systems, and
ings) and the infrared channels, including bands 4, 5, 6, 7, 8 and are as field form in the plain (Fig. 8A and B). The percentage of var-
9 (positive eigenvector loadings). The remaining six PCs can be ex- iance of this ‘vegetation’ PC is only 0.14 and 0.20, showing the
pected to contain information due to the varying spectral response sparse vegetated surface in the study area. Iron oxide minerals
of iron oxides and hydroxyl-bearing minerals (Loughlin, 1991) and have spectral absorption features in the visible to middle infrared
vegetation. from 0.4 to 1.1 lm of the electromagnetic spectrum (Hunt and
By looking for moderate or large eigenvector loadings for bands Ashley, 1979). Vegetation shows absorption features from 0.45 to
1, 2 and 4 in PCs where these loadings are also in opposite sign, we 0.68 lm, and high reflectance in near infrared. It is observed that
can predict that iron oxides can be distinguished by bright pixels in iron oxide minerals have high reflectance in the range of 0.63–
PC4 (Tables 1 and 2). Iron oxide minerals have low reflectance in 0.69 lm, while the range of 0.76–0.90 lm covers higher range of
visible and higher reflectance in near infrared coincide with bands the vegetation red-edge reflectance feature in near infrared, this
1, 2 and band 4 of ASTER data, respectively (Abdelsalam and Stern, characteristic can be used to differentiate iron oxide minerals
2000; Velosky et al., 2003). Electronic processes produce absorp- from vegetation (Crosta and Moore, 1989; Ruiz-Armenta and
tion features in the visible and near infrared radiation (VNIR) Prol-Ledesma, 1998). Hence, bands 2 (0.63–0.69 lm) and 3
(0.4–1.1 lm) due to the presence of transition elements such as (0.78–0.86 lm) of ASTER data include typical features that can be
Fe2+, Fe3+, Mn, Cr, and Ni in the crystal structure of minerals (Hunt, used to separate iron oxide minerals from vegetation. Accordingly,
1977; Hunt and Ashley, 1979). Iron oxides can be mapped as bright vegetation pixels have bright signature in PC5 while iron oxide pix-
pixels in PC4 because of the positive contribution from band 4 els have dark signature (Fig. 8A and B). Al(OH)-bearing minerals
(0.421) and (0.588); while negative contributions from band 1 such as kaolinite, alunite, muscovite and illite show major absorp-
(0.100) and (0.135), band 2 (0.124) and (0.121) for Meiduk tion in bands 5, 6 and 7 (2.14–2.28 lm). Fe, Mg(OH)-bearing min-
and Sarcheshmeh scenes, respectively (Tables 1 and 2). PC4 image erals such as chlorite, as well as carbonates such as calcite and
shows iron oxide minerals that manifested as bright pixels with dolomite have distinctive absorption in bands 8 and 9 (2.29–
circular and semi-circular shapes around known porphyry copper 2.43 lm) of ASTER data (Fig. 2) (Hunt and Ashley, 1979; Mars
deposits and new prospects in the study areas (Fig. 7A and B). Con- and Rowan, 2006). After analyzing the magnitude and sign of the
sidering the PC4 (iron oxide pixels), it has positive great loading of eigenvector loadings and the percentage of variance, it seems that
band 3 (0.417) and (0.391) (Tables 1 and 2, respectively), that re- PC6 and PC7 contain the desired information. PC8 and PC9 are
veals the interruption of vegetation associated with iron oxides noisy and uninformative.
in the PC image as bright pixels. Iron oxides are one of the impor- Eigenvector loadings of bands 4, 5, 6 and 7 in PC6 are (0.519),
tant mineral groups that associated with hydrothermally altered (0.117), (0.486), and (0.204), respectively (Table 1). According
rocks (Sabins, 1999). Iron oxides are create during supergene alter- to Crosta and Moore (1989) and Loughlin (1991) a PC image with
ation and render to characteristic yellowish or reddish color to the moderate to high eigenvector loading for diagnostic reflective
altered rocks that termed gossan (Abdelsalam and Stern, 2000; Xu and absorptive bands of mineral or mineral group with opposite
et al., 2004). Eigenvector loadings for PC5 indicate that PC5 is dom- signs enhances that mineral. If the loading are positive in reflective
inated by vegetation, because vegetation has highly reflectance in band of a mineral the image tone will be bright, and if they are neg-
band 3 and very low in band 2 of ASTER data (Ninomiya, 2003a; ative, the image tone will be dark for the enhanced target mineral.
Xu et al., 2004). The positive loading of band 3 in PC5 (0.573) Thus, eigenvector statistic in each PCA would identify the PC image
and (0.538) and negative loading of band 2 (0.249) and in which the spectral information of mineral under exam is loaded.
(0.225) are shown in Tables 1 and 2 for Meiduk and Sarcheshmeh This information usually represents, in quantitative terms, a very
scenes, respectively. So, Vegetation pixels appear as bright pixels in small fraction of the total information content of the original
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1315

Fig. 7. PC4 image shows iron oxide minerals as bright pixels with circular and semi-
circular shapes that delimited by ellipsoidal polygons in the Meiduk scene (A) and Fig. 8. PC5 image shows vegetation as bright pixels that follow the drainage
Sarcheshmeh scene (B). system. (A) Meiduk scene. (B) Sarcheshmeh scene.

bands, but it is expected that the loaded information indicates in (Fig. 9A). PC6 can be indicator of argilic and phylic alteration zones
the spectral signature of the desired mineral. Considering the mag- because of the magnitude and sign of the eigenvectors loadings of
nitude and sign of the eigenvector loadings of band 4 and also bands 5, 6, 7 in PC6. Analyzing of eigenvector loadings in PC7 indi-
bands 5, 6, 7 in PC6, this is evident that PC image manifests desired cate similar results with small discrepancies in magnitude and sign
information including Al(OH)-bearing minerals as bright pixels of the eigenvector loadings especially in bands 8 (0.638) and 9
1316 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Fig. 10. (A) RGB color composite of PC5, PC6, and PC7 images, showing hydrother-
mal alteration halos around known copper deposits in Meiduk scene. Hydrothermal
alteration halos are delimited by ellipsoidal polygons around the known copper
deposits (highlighted by their names) and identified prospects. Vegetation appears
as red color in the drainage patterns and filed form in the plain. (B) RGB color
composites of PC5, PC6, and PC7 images for Sarcheshmeh scene. Vegetation appears
as light yellow color in the drainage patterns and filed form in the plain. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)

Fig. 9. (A) PC6 image for the Meiduk scene; ellipsoidal polygons delimited the
locations of Al(OH)-bearing minerals as bright pixel. (B) PC7 image for the Meiduk
scene, ellipsoidal polygons delimited the locations of Fe, Mg(OH)-bearing minerals has eigenvector loading with negative sign (0.436) (Table 2).
as bright pixel. According to Yamaguchi and Naito (2003) the reason for these dis-
crepancies can be due to a PCA result which is scene dependent, i.e.
(0.607) (Fe, Mg(OH)-bearing minerals), respectively (Table 1). transform coefficient change from scene to scene. Fig. 10A and B
Therefore, PC7 can be indicator of propylitic alteration zones show RGB color composites of PC5 (Vegetation pixels), PC6
(Fig. 9B). The results of PC images for two ASTER scenes are similar, (Al(OH)-bearing minerals’ pixels), and PC7 (Fe, Mg(OH)-bearing
but the OH-bearing minerals in PC6 and PC7 manifest in dark pix- minerals’ pixels) images of the Meiduk and Sarcheshmeh scenes.
els in the Sarcheshmeh scene because the band 4 (reflective band) The RGB color composites were generated to show vegetation
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1317

covers and hydrothermal alteration zones in study areas. Alter-

ation halos in the Meiduk region are depicted as white to blue color
(Phylic and Argilic zones), and green color (propylitic zone) that
surrounds phyllic and argillic zones, which are easily recognizable
from surrounding rocks. Vegetation covers are appeared as purple
color in the drainage systems and as field form in the plain
(Fig. 10A). The RGB appearances for alteration halos in
Sarcheshmeh scene are manifested in red and yellow colors (Phylic
and Argilic zones) and light blue color (propylitic zone) that
surrounds phyllic and argillic zones, while vegetation covers are
appeared as whitish yellow color (Fig. 10B). In this way, hydrother-
mal alteration zones around known copper deposits and vegeta-
tion are identifiable in the scenes. Some new alteration halos are
also distinguished in Fig. 10A and B, using geology maps as refer-
ence; it seems that a few of them are associated with sedimentary
rocks. Sedimentary rocks (mudstone, shale, claystone, etc.) can be
as erroneous materials in mapping hydrothermal alteration miner-
als due to large amounts of detrital clay minerals in their composi-
tions (Mars and Rowan, 2006). We have delimited the known
copper deposits and new identified alteration halos by ellipsoidal
polygons in the igneous background in Fig. 10A and B.

3.2. Band ratio

Band ratio (BR) transformation is useful for qualitative detec-

tion of hydrothermal alteration minerals. High digital number val-
ues in the scene indicate spectral signatures similar to those of the
particular materials they were designed to map. Ninomiya (2003a)
defined vegetation index for study of distribution vegetation by
ASTER VNIR data. According this definition, the most widely
acknowledged vegetation index is NDVI (Normalized Difference
Vegetation Index), defined as (NIR–red)/(NIR + red), where NIR,
the datum in near infrared, corresponds to ASTER band 3, and
red corresponds to ASTER band 2. It is stable and sensitive enough
for studying vegetation if it is applied to atmospherically corrected
reflectance data. Stabilized vegetation index (StVI) is defined as:
StVI = (band 3/band 2)  (band 1/band 2) (Ninomiya, 2003a). This
index used for distinguishing vegetation in two full ASTER scenes
of the study areas. Fig. 11A and B shows the result of stabilized
vegetation index (StVI) as bright pixel, vegetation highlight in
drainage as linear pattern and in plain as field form. These results
are similar to PC5 of PCA transformation. Velosky et al. (2003) dis-
tinguished the gossan (iron oxide minerals) associated with mas-
sive sulfide mineralization from the host rock by ASTER 4/2 BR
image in the Neoproterozoic Wadi Bidah shear zone, southwestern
Saudi Arabia. Applying ratio of band 4/2 on two subscenes was dis-
tinguished the gossan associated with Mieduk, Sara, Sarcheshmeh,
and Seridune mines as bright pixels (Fig. 12A and B), and again
these identified areas are fully match with PC4 of the PCA transfor-
mation. Previous studies have documented the identification of
specific hydrothermal alteration minerals through the analysis of Fig. 11. ASTER StVI = (band 3/band 2)  (band 1/band 2) image shows vegetation as
bright pixel in Meiduk scene (A), and Sarcheshmeh scene (B).
ASTER BR in the SWIR region. Hewson et al. (2005) suggested BR
of 7/6 for identification of muscovite. We used the BR to highlight
muscovite in the study area because this mineral is a useful tool in
mapping the effects of hydrothermal alteration processes (Van Rui- absorption feature, which is typically centered at 2.20 lm (coin-
tenbeek et al., 2006). Fig. 13A and B shows the output of the BR for cide with ASTER band 6). Propylitic mineral-assemblage reflec-
Meiduk and Sarschesmeh scenes. Hydrothermally altered zones tance spectra are characterized by Fe, Mg-OH absorption features
are shown as bright pixels in the BR 7/6 because muscovite has and CO3 features caused by molecular vibrations in epidote, chlo-
high reflectance in band 7 and low reflectance in band 6. Relative rite and carbonate minerals (Spatz and Wilson, 1995). These
Absorption Band Depths (RBD) consist of RBD5, RBD6, and RBD8 absorption features are situated in the 2.35 lm region (coincide
images have used in this study to delineate argillic and phyllic with ASTER band 8) (Mars and Rowan, 2006). The RBDs have ap-
and propylitic mineral assemblages using ASTER SWIR bands. Arg- plied over two ASTER scenes. RGB color composites were assigned
illic alteration zone is consisted of Kaolinite and alunite that dis- to present the output. In this regard, alteration mineral assem-
plays absorption features at 2.17 lm (coincide with ASTER band blages are demonstrated with different colors, narrow argillic areas
5). Phyllic alteration spectral characteristics include muscovite as bluish green and broad phyllic as yellow color that occupied
and illite reflectance spectra that exhibit an intense Al-OH major parts of the hydrothermal alteration mineral haloes, and
1318 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Fig. 13. ASTER band ratio image of 7/6 shows hydrothermal altered zones
(muscovite), which are located inside ellipsoidal polygons in Meiduk scene (A),
and Sarcheshmeh scene (B).

Fig. 12. ASTER band ratio image of 4/2 shows Iron oxide minerals (Gossan) as bright
pixel in Meiduk and Sara mines (A), and Sarcheshmeh and Seridune mines (B). 3.3. MNF transformation

MNF transformation was performed on SWIR bands of ASTER

propylitic zone as pinkish purple that surrounded outside of these data to detect hydrothermally altered rocks. We focused on the
hydrothermal alteration mineral zones (Fig. 14A and B). The loca- percentage of eginvalues greater than 1 for all of MNF eigenimages,
tion of the alteration haloes are corresponded with highlighted and then RGB color composite applied over results. The percentage
ellipsoidal polygons in the PCA images. However, identified hydro- of eigenvalues for MNF bands is shown in Table 3 for SWIR bands
thermal alteration zones are more recognizable in comparison of ASTER data. According to Boardman and Green (2000) the eigen-
with PCA results. value of each MNF transformed band provides a measure of its
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1319

information content, with progressively noisier bands approaching

eigenvalues near zero. These authors also noted that bands having
low eigenvalues generally have very limited (or no) spatial coher-
ency, again reflecting the dominance of incoherent noise. MNF
eigenimages with values close to 1 contain mostly noise (Jensen,
2005). MNF component images show steadily decreasing image
quality with increasing component number (Chen, 2000). MNF
transformation results derived from SWIR ASTER data indicated
that MNF eigenimages of bands 4, 5 and 6 have eigenvalues per-
cent near to 1 (Table 3). Thus, we excluded the 4th, 5th and 6th
MNF components. We used the remaining eigenimages for produc-
ing RGB color composite. MNF bands 1, 2 and 3 were assigned to
RGB color composites. Fig. 15A and B depicts the output of RGB col-
or composites for two ASTER scenes. In Meiduk scene, hydrother-
mally altered rocks manifest as brownish yellow color. Most of
the rock units’ contacts are recognizable in comparison with the
corresponding geological map (Figs. 15A, and 6). Fig. 15B shows
hydrothermally altered rocks as blue to darkish-blue, and most
of rock units’ contacts are observable in comparison with geologi-
cal map of Sarcheshmeh region (Fig. 5). Because of the erroneous
effect of sedimentary rocks, the known copper deposits and new
identified alteration halos are delimited in the igneous background
by ellipsoidal polygons in the two figures. Identified altered rocks
have different colors in the two study areas probably due to differ-
ent statistical factors for MNF transformed bands, which were as-
signed for generating RGB color composites images (Table 3).
Therefore, altered rocks demonstrated as different colors in Sar-
cheshmeh and Meiduk scenes. The results extracted from MNF
transformation technique for detecting the spatial distribution of
altered rocks were similar to PCA and BR results. This is discernible
that PCA, MNF, and specialized BR transformations have good func-
tion in identifying hydrothermally altered mineral areas and vege-
tation. Some organic materials such as lignin–cellulose have
spectral absorption features centered near 2.10 and 2.30 lm,
which are near the distinctive absorption features of hydrothermal
alteration minerals. The presence of organic materials has affected
the remote detection of hydroxyl-bearing minerals (Van
Ruitenbeek et al., 2006; Mars and Rowan, 2006). Thus, the delinea-
tion of vegetation is paramount in discriminating of hydrother-
mally altered rocks from surrounding area.

3.4. Comparison with previous remote sensing studies and ground

truth

As the literature admitted, a few remote sensing studies were

carried out in the study areas. Tangestani and Moore (2002) used
Thematic Mapper (TM) data for enhancing the alteration patterns
around porphyry intrusive bodies in the Meiduk area. They applied
the Crosta technique, principal component transformation on six
and four TM bands for hydroxyl mapping. Ranjbar et al. (2004) uti-
lized Enhanced Thematic Mapper Plus (ETM+) data for porphyry
Fig. 14. (A) ASTER RBD-ratio image of RBD6, RBD5, RBD8 in RGB. Argillic alteration
zones manifest as bluish green and phyllic alteration zones as yellow color and
copper alteration mapping in the southern part of the Central Ira-
propylitic alteration zones as pinkish purple in Meiduk region. (B) Sarcheshmeh nian Volcanic Belt. Crosta technique performed on selected four
region. Alteration halos associated with known copper deposits and new prospects and six ETM+ bands for enhancing the areas in which the regolith
are highlighted by their names. contains a high proportion of hydroxyl and iron oxide minerals.

Table 3
The percentage of eigenvalues for all of MNF bands extracted from ASTER SWIR data, Meiduk and Sarcheshmeh scenes.

MNF band (Meiduk) Eigenvalues Percent% MNF band (Sarcheshmeh) Eigenvalues Percent%
1 482.6525 85.66 1 457.3547 83.75
2 43.5428 7.72 2 50.3946 9.23
3 15.8743 2.82 3 15.3305 2.81
4 9.6814 1.72 4 10.9961 2.01
5 6.2923 1.11 5 6.5126 1.19
6 5.4405 0.97 6 5.4901 1.01
1320 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Fig. 15. RGB color composite of MNF eigenimages 1, 2, and 3 extracted from ASTER Fig. 16. Field photographs of the study area. (A) Regional view of the open-pit
SWIR bands. (A) Yellow to brownish yellow color areas show hydrothermal altered quarry of Sarcheshmeh porphyry copper mine; (B) view of the phyllic zone; (C)
rocks in Meiduk scene, most of the rock units’ contacts are also observable. (B) view of the argillic zone; (D) close-up of the gossan (iron oxide minerals).
Darkish-blue color areas show hydrothermal altered rocks in Sarcheshmeh scene,
most of the rock units’ contacts are also observable. Known copper deposits
(highlighted by their names) and identified prospects are delimited by ellipsoidal
polygons. (For interpretation of the references to color in this figure legend, the covering the Meiduk porphyry copper mine as well as Sara and Ab-
reader is referred to the web version of this article.) dar copper occurrences. DPCA technique applied on three spectral
bands (4, 5 and 7), PC3 was detected montmorilonite/illite, chlo-
rite, and muscovite minerals. However, the previous investigations
Mars and Rowan (2006) performed logical operator algorithms have not studied many parts of the Central Iranian Volcanic Belt in
based on the Advanced Spaceborne Thermal Emission and Reflec- detail by ASTER remote sensing data and the integration of image
tion Radiometer (ASTER) defined band ratios for regional mapping processing techniques. Therefore, the present study tried to use an
of phyllic and argilic altered rocks in the southern part of the Cen- approach in the image processing techniques to evaluate the AS-
tral Iranian Volcanic Belt. Tangestani et al. (2008) evaluated ASTER TER VNIR and SWIR data for identifying the new prospects of
data for alteration zone enhancement associated with porphyry high-potential alteration zone associated with porphyry copper
copper mineralization in the Meiduk area. Directed Principal Com- mineralization in a regional scale. In this study, the performance
ponent Analysis (DPCA) implemented on a selected subset that of Principal Component Analysis, band ratioing and Minimum
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1321

Noise Fraction transformation has been evaluated for the VNIR and samples in 2170 and 2200 nm and propylitic rock samples in
SWIR subsystems of ASTER data in a regional scale. The image pro- 2350 nm.
cessing techniques indicated the distribution of iron oxides and
vegetation in the VNIR subsystem and hydrothermal alteration ha-
4. Conclusions
los in SWIR subsystem. According to previous remote sensing and
geology studies in the study area (Geological Survey of Iran, 1973;
In this study, ASTER data have been used to identify the hydro-
Tangestani and Moore, 2002; Ranjbar et al., 2004; Mars and Rowan,
thermal alteration zones associated with porphyry copper mineral-
2006; Tangestani et al., 2008), the applied techniques in the pres-
ization, test the data, and image processing techniques for using as
ent study efficiently revealed the alteration halos around known
an exploration tool in the Central Iranian Volcanic Belt with great
copper deposits and identified new prospects. Detected anomalous
potential economic importance and many known porphyry copper
pixels by these techniques are coincided with hydrothermal alter-
deposits, where warranting the exploration of new prospects. Prin-
ation haloes. Results indicate that the integration of the techniques
cipal Component Analysis (PCA), Band ratio, and Minimum Noise
has a great ability to obtain comprehensive and significant infor-
Fraction (MNF) transformation techniques carried out for detailed
mation for the reconnaissance stages of porphyry copper explora-
hydrothermal alteration mapping, resulting in the identification of
tion in a regional scale. This approach in the image processing
high economic-potential zones for copper mineralization. Analysis
techniques can be extrapolated to virgin regions for exploring of
of ASTER level 1 B data after applying Cross-Talk and atmospheric
the new prospect of high-potential copper mineralization zones
corrections, and using the Internal Average Relative Reflection
of the Central Iranian Volcanic Belt and other arid and semi-arid re-
(IARR) method for converting radiance to reflectance, provided
gions of the Earth. The spatial distribution of identified alteration
standard data for hydrothermal alteration mineral mapping. The
zones by the image processing techniques were verified through
image processing techniques have been evaluated over the
in situ inspection. A field reconnaissance was carried out between
Sarcheshmeh and Mieduk mining districts to mapping the prior
10 and 15 December 2010. Geological locations were measured by
known hydrothermal alteration zones. Results of the PCA transfor-
GPS survey with an average accuracy 7 m. Samples for laboratory
mation yielded PC images, which are useful for identifying the spa-
studies were collected through a systematic rock sampling of
tial distribution of specific materials based on their spectral
hydrothermal alteration zones. The field photographs of the hydro-
properties in the VNIR + SWIR bands. PC4, PC5 revealed the distri-
thermal alteration zones are shown in Fig. 16A–D. The mineralogy
bution of iron oxides and vegetation in the VNIR subsystem,
of fine grained samples was studied using the X-ray diffraction
respectively. All anomalous pixels for (OH)-bearing minerals were
(XRD) technique for bulk mineralogy of the hydrothermally altered
detected in PC6 and PC7 images of the SWIR subsystem. It should
rocks. The XRD analyses were implemented on bulk powder using
be noted that the PCA is statistics-based and results may differ in
a X-ray diffractometer D8ADVANCE model. The minerals predom-
the same area with different geologic sizes. Stabilized vegetation
inantly detected in the hydrothermal alteration zones included
index (StVI) and band ratio of 4/2 depicted vegetation and iron oxi-
muscovite, illite and quartz in phyllic zone, kaolinite, montmoril-
des, respectively. The results of band ratio of (7/6) showed the ef-
lonite and quartz in argillic zone, and epidote, chlorite, and quartz
fects of hydrothermal alteration processes that can be utilized as a
in propylitic zone. Spectral reflectance measurements were made
useful tool in mapping high-potential alteration zones. RGB color
using an Analytical Spectral Devices (ASD) filed-portable spec-
composites of RBD5, RBD6, and RBD8 images were useful in
trometer FieldspacÒ HandHeld (HH) model, which records a reflec-
distinguishing of the argillic and phyllic and propylitic mineral
tance spectrum across an overall spectral range of 325–2500 nm
assemblages using ASTER SWIR bands. The results of MNF transfor-
(nm) with a 10 nm individual band width. The measurements were
mation in SWIR were detected hydrothermally altered rocks and
performed at the Remote Sensing laboratory in Technology Univer-
also rock units’ contacts. Results are proven to be effective, and
sity of Malaysia using an artificial light source and contact prob. Al-
in accordance with results of field and laboratory studies. It is
tered rock samples were measured multiple times to get average
shown that the integration of the image processing techniques
spectrum. Fig. 17 indicates the average spectra of collected rock
has great ability to assist economic geologists for the reconnais-
samples from hydrothermal alteration zones. The spectra of phyllic
sance stages of mineral exploration, and can be extrapolated to vir-
rock samples exhibit absorption features in 2200 nm, Argillic rock
gin regions for exploring high-potential copper mineralization
zones.

Acknowledgements

This study was conducted as part of Fundamental Research

Grant Scheme, Ministry of Higher Education Malaysia. We are
grateful to the Universiti Teknologi Malaysia for providing the facil-
ities for this investigation. We also thank reviewers for their com-
ments, which were especially helpful for clarifying certain points
in the manuscript.

References

Abdelsalam, M., Stern, R., 2000. Mapping gossans in arid regions with landsat TM
and SIR-C images, the Beddaho Alteration Zone in northern Eritrea. Journal of
African Earth Sciences 30 (4), 903–916.
Aboelkhair, H., Yoshiki, N., Yasushi, W., Isao, S., 2010. Processing and interpretation
of ASTER TIR data for mapping of rare-metal-enriched albite granitoids in the
Central Eastern Desert of Egypt. Journal of African Earth Sciences 58 (1), 141–
Fig. 17. Laboratory reflectance spectra of altered rock samples, arrows pointed the 151.
maximum absorption. (A) The average spectra of phyllic rock samples has Abrams, M., 2000. The Advanced Spaceborne Thermal Emission and Reflection
maximum absorption in 2200 nm; (B) argillic rock samples in 2170 and 2200 nm; Radiometer (ASTER): data products for the high spatial resolution imager on
(C) propylitic rock samples in 2350 nm. NASA’s Terra platform. International Journal of Remote Sensing 21, 847–859.
1322 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323

Abrams, M., Hook, S.J., 1995. Simulated ASTER data for geologic studies. IEEE Hezarkhani, A., 2006. Hydrothermal evolution of the Sar-Cheshmeh porphyry Cu–
Transactions on Geoscience and remote sensing 33 (3). Mo deposit, Iran: evidence from fluid inclusions. Journal of Asian Earth Sciences
Aftabi, A., Atapour, H., 1997. Geochemical and petrological characteristics of 28, 409–422.
shoshonitic and potassic calcalkaline magmatism at Sarcheshmeh and Hubner, H., 1969a. Geological Map of Iran Sheet No. 5, Southcentral Iran: Tehran,
Dehsiahan porphyry copper deposits, Kerman, Iran. Research Bulletin of National Iranian Oil Company, Scale 1:1000,000.
Isfahan University 9, 127–156 (in Persian). Hubner, H., 1969b. Geological map of Iran sheet no. 6, southeast Iran: Tehran,
Amer, R., Kusky, T., Ghulam, A., 2010. Lithological mapping in the Central National Iranian Oil Company, scale 1:1,000,000.
Eastern Desert of Egypt using ASTER data. Journal of African Earth Sciences Hunt, G., 1977. Spectral signatures of particulate minerals in the visible and near
56, 75–82. infrared. Geophysics 42, 501–513.
Amraie, A., 1991. The Study of Mineralization and Hydrothermal Alteration at Hunt, G.R., Ashley, P., 1979. Spectra of altered rocks in the visible and near infrared.
Meiduk Porphyry Copper Deposit, M.Sc. Thesis. Shiraz University, p. 125. Economic Geology 74, 1613–1629.
Bedell, R.L., 2001. Geological mapping with ASTER satellite: new global satellite Inzana, J., Kusky, T., Higgs, G., Tucker, R., 2003. Supervised classifications of Landsat
data that is a significant leap in remote sensing geologic and alteration TM band ratio images and Landsat TM band ratio image with radar for
mapping. Special Publication Geology Society of Nevada 33, 329–334. geological interpretations of central Madagascar. Journal of African Earth
Bedini, E., Van Der Meer, F., Van Ruitenbeek, F., 2009. Use of HyMap imaging Sciences 37, 59–72.
spectrometer data to map mineralogy in the Rodalquilar caldera, southeast Jensen, J.R., 2005. Introductory Digital Image Processing. Person Prentice Hall, Upper
Spain. International Journal of Remote Sensing 30 (2), 327–348. Saddle River, pp. 307 and 444.
Ben-Dor, E., Kruse, F.A., lefkoff, A.B., Banin, A., 1995. Comparison of three calibration Kargi, H., 2007. Principal components analysis for borate mapping. International
techniques for utilization of GER 63-channel aircraft scanner data of Makhtesh Journal of Remote Sensing 28 (8), 1805–1817.
Ramon, Nega, Israel. International Journal of Rock Mechanics and Mining Khan, S.D., Mahmood, K., 2008. The application of remote sensing techniques to the
science and Geomechanics Abstract 32, 164A. study of ophiolites. Earth-Science Reviews 89, 135–143.
Berberian, M., King, G.C., 1981. Towards a paleogeography and tectonic evolution of Loughlin, W.P., 1991. Principal components analysis for alteration mapping.
Iran. Canadian Journal of Earth Sciences 18, 210–265. Photogrammetric Engineering and Remote Sensing 57, 1163–1169.
Boardman, J.W., Green, R.O., 2000. Exploring the Spectral Variability of the Earth as Lowell, J.D., Guilbert, J.M., 1970. Lateral and vertical alteration-mineralization
Measured by AVIRIS in 1999. Summaries of the Ninth Annual JPL Airborne zoning in porphyry ore deposits. Economic Geology and the Bulletin of the
Geosciences Workshop. Jet Propulsion Laboratory Special Publication 18, p. 10. Society of Economic Geologists 65 (4), 373–408.
Boardman, J.W., Kruse, F.A., Green, R.O., 1995. Mapping Target Signatures via Partial Mars, J.C., Rowan, L.C., 2006. Regional mapping of phyllic- and argillic-altered rocks
Unmixing of AVIRIS data. Summaries of the 5nd Annual JPL Airborne Geoscience in the Zagros magmatic arc, Iran, using Advanced Spaceborne Thermal Emission
Workshop, JPL Publication 95-1, vol. 1, pp. 23–26. and Reflection Radiometer (ASTER) data and logical operator algorithms.
Boomeri, M., Kazuo, N., David, R.L., 2009. The Meiduk porphyry Cu deposit, Kerman, Geosphere 2 (3), 161–186.
Iran: a geochemical analysis of the potassic zone including halogen element Massironi, M.L., Bertoldi, P., Calafa, D., Visona, A., Bistacchi, C., Giardino, A., Schiavo,
systematics related to Cu mineralization processes. Journal of Geochemical B., 2008. Interpretation and processing of ASTER data for geological mapping
Exploration 103, 17–29. and granitoids detection in the Saghro massif (eastern Anti-Atlas, Morocco).
Chang, Q., Jing, L., Panahi, A., 2006. Principal component analysis with optimum Geosphere 4 (4), 736–759.
order sample correlation coefficient for image enhancement. International Modarres, R., Da Silva, V.P.R., 2007. Rainfall trends in arid and semi-arid regions of
Journal of Remote Sensing 27 (16), 3387–3401. Iran. Journal of Arid Environments 70, 344–355.
Chen, C.M., 2000. Comparison of principal component analysis and Minimum Noise Moghtaderi, A., Moore, F., Mohammadzadeh, A., 2007. The application of advanced
Fraction transformation for reducing the dimensionality of hyperspectral space-borne thermal emission and reflection (ASTER) radiometer data in the
imagery. Geographical Research 33, 163–178. detection of alteration in the Chadormalu paleocrater, Bafq region, Central Iran.
Clark, R.N., Swayze, G.A., Gallagher, A., King, T.V.V., Calvin, W.M., 1993b. The US Journal of Asian Earth Sciences 30, 238–252.
Geological Survey, Digital Spectral Library: Version 1: 0.2 to 3.0 microns: US Moore, F., Rastmanesh, F., Asady, H., Modabberi, S., 2008. Mapping mineralogical
Geological Survey Open File Report 93-592, 1340 p. <http://speclab.cr.usgs.gov> alteration using principal component analysis and matched filter processing in
(August 1999). Takab area, north-west Iran, from ASTER data. International Journal of Remote
Crosta, A., Moore, J. McM., 1989. Enhancement of Landsat Thematic Mapper imagery Sensing 29 (10), 2851–2867.
for residual soil mapping in SW Minais Gerais State, Brazil: a prospecting case Ninomiya, Y., 2003a. A stabilized vegetation index and several mineralogic indices
history in Greenstone belt terrain. In: Proceedings of the 7th ERIM Thematic defined for ASTER VNIR and SWIR data. In: Proceedings of the IEEE 2003
Conference: Remote Sensing for Exploration Geology, pp. 1173–1187. International Geoscience and Remote Sensing Symposium (IGARSS’03), vol. 3,
Crosta, A.P., Souza Filho, C.R., Azevedo, F., Brodie, C., 2003. Targeting key alteration Toulouse, France, 21–25 July 2003, pp. 1552–1554.
minerals in epithermal deposits in Patagonia, Argentina, using ASTER imagery Ninomiya, Y., 2003b. Advanced remote lithologic mapping in ophiolite zone with
and principal component analysis. International Journal of Remote sensing 24, ASTER multispectral thermal infrared data. In: Proceedings of the IEEE 2003
4233–4240. International Geoscience and Remote Sensing Symposium (IGARSS’03), vol. 3,
Crowley, J.K., Brickey, D.W., Rowan, L.C., 1989. Airborne imaging spectrometer data Toulouse, France, 21–25 July 2003, pp. 1561–1563.
of the Ruby Mountains, Montana: mineral discrimination using relative Pour, B.A., Hashim, M., Marghany, M., 2011. Using spectral mapping techniques on
absorption band-depth images. Remote Sensing of Environment 29 (2), 121– short wave infrared bands of ASTER remote sensing data for alteration mineral
134. mapping in SE Iran. International Journal of the Physical Sciences 6 (4), 917–
Di Tommaso, I.D., Rubinstein, N., 2007. Hydrothermal alteration mapping using 929.
ASTER data in the Infiernillo porphyry deposit, Argentina. Ore Geology Reviews Ranjbar, H., Honarmand, M., Mozafari, Z., 2004. Application of the Crosta technique
32, 275–290. for porphyry copper alteration mapping, using ETM+ data in the southern part
Fujisada, H., 1995. Design and performance of ASTER instrument. In: Breckinridge, of the Iranian volcanic sedimentary belt. Journal of Asian Earth Sciences 24,
J.B. (Ed.), Proceedings of International Society of Optical Engineering, vol. 2583, 237–243.
pp. 16–25. Rockwell, B.W., Hofstra, A.H., 2008. Identification of quartz and carbonate minerals
Gabr, S., Ghulam, A., Kusky, T., 2010. Detecting areas of high-potential gold across northern Nevada using ASTER thermal infrared emissivity data
mineralization using ASTER data. Ore Geology Reviews 38, 59–69. Implications for geologic mapping and mineral resource investigations in
Gad, S., Kusky, T., 2007. ASTER spectral ratioing for lithological mapping in the well-studied and frontier areas. Geosphere 4 (1), 218–246.
Arabian–Nubian shield, the Neoproterozoic Wadi Kid area, Sinai, Egypt. Rowan, L.C., Mars, J.C., 2003. Lithologic mapping in the Mountain Pass, California
Gondwana Research 11, 326–335. area using Advanced Spaceborne Thermal Emission and Reflection Radiometer
Geological Survey of Iran, 1973. Exploration for ore deposits in Kerman region. (ASTER) data. Remote Sensing of Environment 84, 350–366.
Report No. Yu/53. Rowan, L.C., Hook, S.J., Abrams, M.J., Mars, J.C., 2003. Mapping hydrothermally
Gersman, R., Ben-Dor, E., Beyth, M., Avigad, D., Abraha, M., Kibreba, A., 2008. altered rocks at Cuprite, Nevada, using the Advanced Spaceborne Thermal
Mapping of hydrothermally altered rocks by the EO-1 Hyperrion sensor, Emission and Reflection Radiometer (ASTER), a new satellite-imaging system.
northern Danakil, Eritrea. International Journal of Remote Sensing 29 (13), Economic Geology 98 (5), 1019–1027.
3911–3936. Rowan, L.C., Robert, G.S., John, C., 2006. Distribution of hydrothermally altered rocks
Green, A.A., Berman, M., Switzer, P., Craig, M.D., 1988. A transformation for in the Reko Diq, Pakistan mineralized area based on spectral analysis of ASTER
ordering multispectral data in terms of image quality with implications for data. Remote Sensing of Environment 104, 74–87.
noise removal. IEEE Transactions on Geoscience and Remote Sensing 26 (1), Ruiz-Armenta, J.R., Prol-Ledesma, R.M., 1998. Techniques for enhancing the spectral
65–74. response of hydrothermal alteration minerals in Thematic Mapper images of
Hassan-Nezhad, A.A., Moore, F., 2006. A stable isotope and fluid inclusion study of Central Mexico. International Journal of Remote Sensing 19, 1981–2000.
the Qaleh-Zari Cu–Au–Ag deposit, Khorasan Province, Iran. Journal of Asian Sabins, F.F., 1999. Remote sensing for mineral exploration. Ore Geology Reviews 14,
Earth Sciences 27, 805–818. 157–183.
Hassanzadeh, J., 1993. Metallogenic and tectonomagmatic events in the SE sector of Shafiei, B., 2010. Lead isotope signatures of the igneous rocks and porphyry copper
the Cenozoic active continental margin of central Iran (Shahr e Babak area, deposits from the Kerman Cenozoic magmatic arc (SE Iran), and their
Keman Province). Ph.D. thesis: University of California, Los Angeles, Los Angeles, magmatic-metallogenetic implications. Ore Geology Reviews 37 (1–2), 27–36.
p. 204. Shahabpour, J., 2005. Tectonic evolution of the orogenic belt in the region located
Hewson, R.D., Cudahy, T.J., Mizuhiko, S., Ueda, K., Mauger, A.J., 2005. Seamless between Kerman and Neyriz. Journal of Asian Earth Sciences 24, 405–417.
geological map generation using ASTER in the Broken Hill–Curnamona province Shahabpour, J., 2007. Island-arc affinity of the Central Iranian Volcanic Belt. Journal
of Australia. Remote Sensing of Environment 99, 159–172. of Asian Earth Sciences 30, 652–665.
 A.B. Pour, M. Hashim / Journal of Asian Earth Sciences 42 (2011) 1309–1323 1323

Singh, A., Harrison, A., 1985. Standardized principal components. International Saudi Arabia, inferences from orbital remote sensing and field studies.
Journal of Remote Sensing 6, 883–896. Precambrian Research 123 (2–4), 235–247.
Spatz, D.M., Wilson, R.T., 1995, Remote sensing characteristics of porphyry copper Waterman, G.C., Hamilton, R.L., 1975. The Sarcheshmeh porphyry copper deposit.
systems, western America Cordillera. In: Pierce, F.W., Bolm, J.G. (Eds.), Arizona Economic Geology 70, 568–576.
Geological Society Digest, vol. 20, pp. 94–108. Xu, Y., Qizhong, L., Yun, S., Lu, W., 2004. Extraction Mechanism of Alteration Zones
Tangestani, M.H., Moore, F., 2002. Porphyry copper alteration mapping at the using ASTER Imagery. 0-7803-8742-2/04 (c) 2004 IEEE.
Meiduk area, Iran. International Journal of Remote Sensing 23 (22), 4815–4825. Yamaguchi, Y., Naito, C., 2003. Spectral indices for lithologic discrimination and
Tangestani, M.H., Mazhari, N., Ager, B., Moore, F., 2008. Evaluating advance mapping by using the ASTER SWIR bands. International Journal of Remote
spaceborne thermal emission and reflection radiometer (ASTER) data for Sensing 24 (22), 4311–4323.
alteration zone enhancement in a semi-arid area, northern shahr-e-Babak, SE Yamaguchi, Y.I., Fujisada, H., Kudoh, M., Kawakami, T., Tsu, H., Kahle, A.B., Pniel, M.,
Iran. International Journal of Remote Sensing 29 (10), 2833–2850. 1999. ASTER instrument characterization and operation scnorio. Advanced
Van Ruitenbeek, F.J.A., Pravesh, D., van der Meer, F.D., Cudahy, T., van der Meijde, M., Space Research 23 (8), 1415–1424.
Hale, M., 2006. Mapping white micas and their absorption wavelengths using Zhang, X., Micha, P., Norman, D., 2007. Lithologic and mineral information
hyperspectral band ratios. Remote Sensing of Environment 102, 211–222. extraction for gold exploration using ASTER data in the south Chocolate
Velosky, J.C., Stern, R.J., Johnson, P.R., 2003. Geological control of massive sulfide Mountains (California). Journal of Photogrammetry and Remote Sensing 62,
mineralization in the Neoproterozoic Wadi Bidah shear zone, southwestern 271–282.

View publication stats