Accepted Manuscript Title: Hydrothermal alteration mapping using Landsat-8 data, Sar Cheshmeh copper mining district, SE Iran Author: Amin Beiranvand Pour Mazlan Hashim PII: DOI: Reference:
S1658-3655(14)00129-0 http://dx.doi.org/doi:10.1016/j.jtusci.2014.11.008 JTUSCI 121
To appear in: Received date: Revised date: Accepted date:
11-9-2014 2-10-2014 14-11-2014
Please cite this article as: A.B. Pour, Hydrothermal alteration mapping using Landsat-8 data, Sar Cheshmeh copper mining district, SE Iran, Journal of Taibah University for Science (2014), http://dx.doi.org/10.1016/j.jtusci.2014.11.008 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Hydrothermal alteration mapping using Landsat-8 data, Sar Cheshmeh copper mining district, SE Iran
ip t
Amin Beiranvand Pour*, Mazlan Hashim Institute of Geospatial Science & Technology (INSTeG), Universiti Teknologi Malaysia, 81310
us
cr
UTM Skudai, Johor Bahru, Malaysia
an
Abstract
This study presents the applicability of recently launched Landsat-8 data for hydrothermal
M
alteration and lithological mapping aim at porphyry copper exploration in arid and semi-arid regions. Sar Cheshmeh copper mining district in the southeastern part of the Urumieh-Dokhtar
d
volcanic belt, SE Iran has been selected as a case study. Several Red-green-Blue (RGB) color
te
combination images and specialized band ratios were developed using Landsat-8 bands. Band
Ac ce p
ratios derived from image spectra (4/2, 6/7, 5 and 10 in RGB) allow the identification of altered rocks, lithological units and vegetation at a regional scale. Analytical imaging and geophysics developed hyperspectral analysis processing methods were applied to Landsat-8 bands to identify alteration zone associated with known porphyry copper deposits. Mixture Tuned Matched Filtering (MTMF) method was used to detect alteration zones associated with known porphyry copper deposits in the study area. Fieldwork, pervious remote sensing studies and laboratory analysis were utilized to verify the image processing results derived from Landsat-8 bands. It is concluded that Landsat-8 bands especially bands 2 and 4 in visible and near-infrared, 6 and 7 in shortwave infrared, and 10 in thermal infrared contain useful spectral information for porphyry copper exploration purposes. Moreover, thermal infrared bands of Landsat-8 1 Page 1 of 32
significantly improved the quality and availability of thermal infrared remote sensing data for lithological mapping. The achievements of this investigation should have considerable implications for geologists to utilize Landsat-8 Operational Land Imager (OLI)/ Thermal Infrared
ip t
Sensor (TIRS) data for porphyry copper exploration and geological purposes in the future.
cr
Key words: Landsat-8 data, Operational Land Imager, Thermal Infrared Sensor, Hydrothermal alteration mapping,
us
Porphyry copper exploration.
*Corresponding author. Tel: +607-5530666; Fax: +607-5531174; Email address:
[email protected];
M
an
[email protected];
[email protected],
[email protected].
1. Introduction
d
Remote sensing technology has been used in diverse aspects of Earth sciences, geography,
te
archeology and environmental sciences. New generations of advanced remote sensing data have
Ac ce p
been used by Earth scientists over last decades. They have focused on global experiences in environmental geology, mineral and hydrocarbon exploration. In the initial stage of remote sensing technology development (1970s), geological mapping and mineral exploration were among the most prominent applications [1, 2, 3, 4]. Multispectral and hyperspectral remote sensing sensors were used for geological applications, ranging from a few spectral bands to more than 100 contiguous bands, covering the visible to the shortwave infrared regions of the electromagnetic spectrum [5, 6, 7,8, 9, 10, 11,12, 13, 14, 15, 16, 17,18, 19, 20,21]. The Landsat satellites era that began in 1972 will become a nearly 42-year global land record with the successful launch and operation of the Landsat Data Continuity Mission (LDCM). Two generations of Landsat satellites has been launched by National Aeronautics and Space 2 Page 2 of 32
Administration (NASA) and the U.S. Geological Survey (USGS). The first generation (Landsats 1, 2, and 3) operated from 1972 to 1985 and is essentially replaced by the second generation (Landsats 4, 5, and 7), which began in 1982 and continues to the present. Landsat 6 of the second
ip t
generation was launched in 1993, but failed to reach orbit [22]. The LDCM is a partnership formed between the NASA and the USGS to place the next Landsat satellite in orbit.
cr
The Advanced Land Imager (ALI) sensor was launched on 21 November of 2000 as archetype
us
for the next production Landsat satellites, the multispectral characteristics maintains to Enhanced Thematic Mapper Plus (ETM+) sensor on Landsat-7 with a spatial resolution of 30 m, but the
an
swath width is 37 km [23, 24, 25, 26].
Landsat-8 was launched on 4 February 2013 from Vandenberg Air Force Base in California. It is
M
an American Earth observation satellite and the eighth satellite in the Landsat program. Landsat-
d
8 joins Landsat-7 on-orbit, providing increased coverage of the Earth’s surface. It is in the form
te
of free-flyer spacecraft carrying a two-sensor payload, the Operational Land Imager (OLI) and the Thermal Infrared Sensor (TIRS). These two instruments collect image data for nine visible,
Ac ce p
near-infrared, shortwave infrared bands and two longwave thermal bands. They have high signal to noise (SNR) radiometer performance, enabling 12-bit quantization of data allowing for more bits for better land-cover characterization. Landsat-8 provides moderate-resolution imagery, from 15 meters to 100 meters of Earth’s surface and polar regions [27, 28]. Landsat-8 data have been distributed to the general public on nondiscriminatory basis at no cost to the user. The data can be easily downloaded from the (http://earthexplorer.usgs.gov and http://glovis.usgs.gov/) online linkages. This revolution has allowed scientists to detect changes over time to our planet and has enabled a host of applications based on Landsat measurements to be developed by researchers, the private
3 Page 3 of 32
sector, and state, local, and tribal governments. The performance characteristics of the ALI, ETM+ and Landsat-8 are shown in Table 1. Landsat-8 OLI/TIRS data can be used to monitor variety of earth-based and atmospheric phenomenon, including agricultural monitoring;
ip t
geological mapping; evapotranspiration; cloud detection and analysis; mapping heat fluxes from cities; monitoring air quality; monitoring volcanic activity; monitoring the rain forests; biomass
cr
burning; industrial thermal pollution in the atmosphere, rivers and lakes; monitoring/tracking
us
material transport in lakes and coastal regions; identifying insect breeding areas and applications that will ultimately arise in the future as a result of global warming and climate change.
an
This study evaluates the capability of Landsat-8 bands for mapping hydrothermal alteration area and lithological units associated with porphyry copper deposits in arid and semi-arid regions. Sar
M
Cheshmeh copper mining district has been selected as a case study, which is located in the
d
southeastern part of the Urumieh-Dokhtar volcanic belt, SE Iran (Figs. 1 and 2), where Cu and
te
Mo are actively being mined. Several remote sensing studies have been conducted by authors in Cheshmeh copper mining district using ASTER, ALI and Hyperion data during recent years. In
Ac ce p
the previous study, we applied many image processing techniques such as principal component analysis, band ratioing and spectral mapping methods to the ASTER, ALI and Hyperion to detect alteration zones associated with porphyry copper deposits in the study area. However, freely available Landsat-8 imagery has not been evaluated for porphyry copper exploration in this area and other arid and semi-arid regions. Hence, more investigation is required to test the application of Landsat-8 data for locating potential porphyry copper deposits in and semi-arid regions. The objectives of this study are: (1) to evaluate Landsat-8 spectral bands for detecting hydrothermal alteration minerals and rock units associated with porphyry copper deposits in arid and semi-arid regions; (2) to developed specialized band ratios for porphyry copper exploration purposes using
4 Page 4 of 32
Landsat-8 bands; (3) to test mixture tuned matched filtering spectral mapping method on
ip t
Landsat-8 OLI bands for alteration zone detection.
cr
2. Geology of the study area
us
The Sar Cheshmeh porphyry copper deposit (55◦ 52′ 20″ E, 29◦ 58′ 40″ N) is located 60 Km southwest of Kerman city in Kerman province, southeastern part of Iran. Figure 2 shows
an
geological map of the Sar Cheshmeh area [29, 30]. The deposit is within a belt of Eocene volcanic rocks and Oligo-Miocene subvolcanic granitoid rocks. The oldest host rocks at the Sar
M
Cheshmeh porphyry copper deposit belong to an Eocene volcanogenic complex, also known as the Sar Cheshmeh complex [31]. The complex consists of pyroxene trachybasalt, pyroxene
d
trachyandesite of potassic and shoshonitic affinity [32], less abundant andesite and rare
te
occurrences of agglomerate, tuff and tuffaceous sandstone. These were intruded by a complex
Ac ce p
series of Oligo-Miocene granitoid intrusive phases such as quartz diorite, quartz monzonite and granodiorite. The granitoid rocks are cut by a series of intramineral hornblende porphyry, feldspar porphyry and biotite porphyry dykes. Hydrothermal alteration and mineralization at Sar Cheshmeh occur as stockworks, which were broadly synchronous with the granitoid intrusives in their emplacement. Early hydrothermal alteration was predominantly potassic and propylitic, but followed later by phyllic, silicic and argillic alteration [33]. The Sar Cheshmeh deposit is estimated to contain approximately 1.200 million tons of 1.2 percent copper with significant amounts of molybdenum (0.03 percent) and gold (0.01 percent) [34]. Several remote sensing studies have been conducted in this region during recent years [35, 36, 37, 38]. 5 Page 5 of 32
3. Material and methods 3.1 Landsat-8 OLI/ TIRS Data A cloud-free level 1T (terrain corrected) Landsat-8 image LC81600392013135LGN01
ip t
(Path/Row 160/39) was obtained through the U.S. Geological Survey Earth Resources
cr
Observation and Science Center (EROS) (http://earthexplorer.usgs.gov). It was acquired on May 15, 2013 for the Sar Cheshmeh area. The image map projection is UTM zone 40 North (Polar
us
Stereographic for Antractica) using the WGS-84 datum. The Operational Land Imager (OLI) feature two additional spectral channels with advanced measurement capabilities include: a deep-
an
blue band for coastal water and aerosol studies (band 1, 0.433-0.453 μm, 30 m pixel size), and a
M
band for cirrus cloud detection (band 9, 1.360-1.390 μm, 30 m pixel size). The Thermal Infrared Sensor (TIRS) collects data in two long wavelength thermal bands (band 10, 10.30-11.30 μm,
te
registered with OLI data (Table 1).
d
100 m pixel size; band 11, 11.50-12.50 μm, 100 m pixel size), which have already been co-
Ac ce p
3.2 Preprocessing of Landsat-8 OLI/ TIRS Data Landsat-8 image of target site was processed using the ENVI (Environment for Visualizing Images) version 4.8 software package. Landsat-8 data were converted to surface reflectance using the Internal Average Relative Reflection (IARR) method [39]. IARR reflectance technique is recommended for mineralogical mapping as a preferred calibration technique, which does not require the prior knowledge of samples collected from the field. During the atmospheric correction, raw radiance data from imaging spectrometer is re-scaled to reflectance data. Therefore, all spectra are shifted to nearly the same albedo. The resultant spectra can be compared with the reflectance spectra of the laboratory or filed spectra, directly. The
6 Page 6 of 32
panchromatic and cirrus cloud (band 9) bands have not been used in this study. Thermal Atmospheric Correction was performed on TIR bands, which is based on a normalized pixel regression method [40]. The 90 m resolution TIR bands were also re-sampled to correspond to
ip t
the 30-m spatial dimensions for some image processing applications. Nearest neighbor re-
us
cr
sampling method was applied to preserve the original pixel values in the re-sampled images.
an
3.3 Image processing methods
The ability to discriminate between hydrothermaly altered and unaltered rocks are considerable
M
in mineral exploration studies. In the region of solar reflected light (0.325 to 2.5 μm), many minerals demonstrate diagnostic absorption features due to vibrational overtones, electronic
d
transition, charge transfer and conduction processes. Sericitically-altered rocks typically contain
te
sericite, a fine-grained form of muscovite that has a distinct Al-OH absorption feature at 2.2 μm
Ac ce p
and a less intense absorption feature at 2.35 μm (Figure 3 A). Kaolinite and alunite are typical constituents of advanced argillic alteration that exhibit Al-OH 2.165 μm and 2.2 μm absorption features (Figure 3 A). Propylitically-altered rocks typically contain varying amounts of chlorite, epidote and calcite, which exhibit Fe, Mg-O-H and CO3 2.31–2.33 μm absorption features (Figure 3 A). Iron oxide/hydroxide minerals such as limonite, jarosite and hematite tend to have spectral absorption features in the visible to middle infrared from 0.4 to 1.1 μm of the electromagnetic spectrum (Figure 3 B) [41, 42]. Hydrothermal silica minerals typically consist of quartz, opal and chalcedony. TIR emissivity spectra illustrate that quartz and opal contain a prominent restrahlen feature in the 9.1 μm region [43, 44].
7 Page 7 of 32
The reflectance of samples from porphyry copper deposits exhibited strong absorption features at 2200 and 2170 nm. Reflectance curves exhibited similarities to spectra of muscovite (strong absorption feature) and kaolinite and minor signatures of alunite. Samples from surrounding
ip t
areas showed strong absorption features at 2350 nm. This feature is attributed to the existence of epidote and chlorite in the altered rocks. Absorption features at 2200, 2170 and 2350 nm were
cr
found to be the most diagnostic spectral feature of the sericitically, argillically and propylitically-
us
altered rocks associated with porphyry copper mineralization, respectively.
Based on laboratory spectra of the minerals related with hydrothermal alteration and lithological
an
units several Red-green-Blue (RGB) color combination images and band ratios were created using Landsat-8 bands in this research. Different Red-green-Blue (RGB) color combination
M
images were applied for enhancing the hydrothermally altered rocks and lithological units at
d
regional scale. Band ratioing is a technique where the digital number value of one band is
te
divided by the digital number value of another band. Band ratios are very useful for highlighting certain features or materials that cannot be seen in the raw bands [45].
Ac ce p
Spectral mapping methods aim at extracting individual mineral species from a mixed pixel spectrum, in theory providing geologist with the capability to map mineral surface composition. Although these image processing methods have been mostly applied on hyperspectral data, they can also be applicable logically to multispectral data. With these image processing methods, pixels that have mixed spectral signatures will be extractable and can be separated from the undesirable background. Thus, mineral abundance maps can be produced free of diluting effects of surrounding environment. Analytical imaging and geophysics (AIG)-developed hyperspectral analysis processing methods are applied on Landsat-8 bands for mapping hydroxide and carbonate minerals assemblages in hydrothermal alteration zones. AIG approach for analysis of
8 Page 8 of 32
hyperspectral data are implemented and documented within the ENVI software system. Data are analyzed using AIG approach to determine unique spectral end-members and their spatial distribution, abundances and producing detailed mineral maps.
ip t
The analysis approach consists of the following steps: (1) spectral compression, noise suppression and dimensionality reduction using the Minimum Noise Fraction (MNF)
cr
transformation [46, 47]; (2) spatial data reduction using the Pixel Purity Index (PPI) (Boardman,
us
1995); (3) extraction of end-member spectra using n-Dimensional Visualizer [46]; (4) identification of end-member spectra using visual inspection, automated identification and
an
spectral library comparisons [48]; and finally (5) production of material maps using a variety of spectral mapping methods such as Spectral Angle Mapper (SAM), Linear Spectral Unmixing
d
Fitting (SFF) and Binary Encoding (BE).
M
(LSU), Matched Filtering (MF), Mixture Tuned Matched Filtering (MTMF), Spectral Feature
te
Mixture Tuned Matched Filtering (MTMF) method was selected and applied to nine visible, near-infrared, shortwave infrared bands of Landsat-8 to identify alteration zone associated with
Ac ce p
known porphyry copper deposits in the study area. MTMF is a combination of the best parts of the linear spectral mixing model and the statistical matched filter model while avoiding the drawbacks of each parent method. From matched filtering it inherits the advantage of its ability to map a single known target without knowing the other background end-member signatures, unlike traditional spectra mixture modeling. From spectral mixture modeling it inherits the leverage arising from the mixed pixel model, the constraints on feasibility including the unit-sum and positivity requirements, unlike the matched filter which does not employ these fundamental facts. As a result MTMF can outperform either method, especially in cases of subtle, sub-pixel occurrences [48].
9 Page 9 of 32
The pervious remote sensing studies, filed observations, Global Positioning System (GPS) survey, X-ray diffraction (XRD) analysis, field spectral reflectance measurements and regional geology map of the study area were used to verify the output results derived from image
ip t
processing methods. Additionally, the Root Mean Square Error (RMSE) analysis was performed for 50 alteration mapped pixel points derived of from the image possessing results and compared
( Preal,i Pestimated ,i ) 2 n
us
i 1
an
RMSE
n
cr
with real points on the ground obtained by GPS survey.
(1.1)
M
where Preal is real points on the ground and Pestimated is alteration mapped pixel points at point i.
Ac ce p
te
d
The estimated RMSE for selected points in this study was 0.7865.
4. Results and discussion
The first spectral band of Landsat-8 (0.433-0.453 μm) is designed as a deep-blue band for coastal water and aerosol studies, so it cannot be used to detect geological features. Band 1 excluded from RGB color combination image in this section. A single RGB image was produced for visible bands (2, 3 and 4) of Landsat-8 data. Band 2 has positioned in the blue (0.450-0.515 μm), band 3 in green (0.525-0.600 μm) and band 4 in red (0.630-0.680 μm) regions of the electromagnetic spectrum. Natural RGB color combination image has been assigned to bands 4, 10 Page 10 of 32
3 and 2 for full scene of the image. Geological features and geomorphological framework are distinguishable at a regional scale. Textural characteristics of the igneous rocks can be discriminated from sedimentary rock in the scene. Structural features and sedimentary texture of
ip t
rocks in the scene are easily recognizable using this natural RGB color combination of the
cr
visible bands.
RGB color combination image was allocated to near infrared (band 5: 0.845-0.885 μm) and short
us
wavelength infrared bands (band 6:1.560-1.660 μm , band 7: 2.100-2.300 μm) of Landsat-8 data,
an
respectively. Geological features, including textural characteristics of the igneous and sedimentary rocks, structural features and vegetation are detected at a regional scale. Vegetated
M
areas are appeared as light red color in the scene. Vegetation shows absorption features from 0.45 to 0.68 μm, and high reflectance in near infrared from 0.7 to 1.2 μm [49]. So, vegetated
d
areas are more visible in the resultant RGB image having near infrared band (band 5). Color and
te
textural features of the igneous and sedimentary rocks are more robust using this RGB color
Ac ce p
composite. Moreover, hydrothermal altered rocks are also recognizable as yellow color area within the belt of crystalline igneous rocks in the Sar Cheshmeh copper mining district. Clay and carbonate minerals have absorption features from 2.1 to 2.4 μm (band 7 of Landsat-8) and reflectance from 1.55 to 1.75 μm (band 6 of Landsat-8) [50]. Two thermal infrared bands (bands 10 and 11) of Landsat-8 have spectral coverage in 10.3011.30 μm and 11.50-12.50 μm, respectively. The energy measured by TIR bands from the Earth’s surface is a function of temperature as well as the emissivity of the target, which is dependent on its chemistry and texture [50]. TIR bands of Landsat-8 have improved the quality and applicability of the Landsat data in a variety of earth-based and atmospheric phenomenon [23]. Silicate minerals exhibit spectral features in the TIR. The silica emissivity curve shows 11 Page 11 of 32
significant variation in 8.5 μm to 9.30 μm and 10.30 to 11.70 μm. There are clear minima in 8.5 μm to 9.30 μm while higher emissivity values can be seen in 10.30 to 11.70 μm [50]. RGB color combination image can be produced for band 10 (10.30-11.30 μm), 11 (11.50-12.50 μm) and 7
ip t
(2.100-2.300 μm) of Landsat-8 at a regional scale. Rocks with high emissivity value attributed to high silicate minerals in their composition are manifested as red color in the image. On the other
cr
hand, rocks with moderate and low emissivity value are appeared as pink and blue colors,
us
respectively. Band 7 has been selected for RGB color combination image as representative of rocks absorption features in SWIR region due to Al–OH, Fe, Mg–OH, Si–O–H and CO3 in their
an
compositions. Hence, blue hue areas contain low silicate minerals rocks.
M
Accordingly, for identification of hydrothermal alteration minerals associated with porphyry copper mineralization using Landsat-8 bands, two band ratios have been developed based on
d
laboratory spectra of alteration minerals. Mapping iron oxides is carried out using bands 2 and 4
te
because iron oxide/hydroxide minerals such as hematite, jarosite and limonite have high
Ac ce p
reflectance within 0.63 to 0.69 μm (the equivalent to ETM+ band 3) and high absorption within 0.45 to 0.52 μm (the equivalent to ETM+ band 1). The analysis to map clay and carbonate minerals must incorporate bands 6 and 7 attributed to high reflectance in the range of 1.55 to 1.75 μm and high absorption in 2.08 to 2.35 μm that correspond with ETM+ bands 5 and 7, respectively.
Band ratios derived from image spectra (4/2, 6/7, 5 and 10 in RGB) allows the identification of altered rocks, lithological units and vegetation (Figs. 4 and 5). The alteration minerals (hydrothermally altered rocks) are outlined in the images where they appear as yellow color around known and mined porphyry copper deposits, which are more visible in figure 5. The known and mined copper deposits are highlighted by their names in the Figures. The boundary 12 Page 12 of 32
between sedimentary (Neogene redbed agglomerate and conglomerate) and igneous rocks (Eocene-Oligocene volcanic rocks, Lower Eocene tuff and volcanic rocks) units are also delimited in the resultant images. Vegetation is manifested as red and purple colors in the
ip t
drainage system and background of both scenes (Figs. 4 and 5). According to pervious remote sensing study using ASTER, ALI, ETM+ and Hyperion data in the study area these two band
cr
ratios derived from Landsat-8 bands efficiently identified hydrothermal alteration rocks and
us
lithological units. However, geological features are more visible in 4/2, 6/7, 10 band ratio image due to the existence of TIR band (band 10) in the RGB color combination. As above mentioned,
an
the emissivity of the geological target in the TIR region is dependent on its chemistry and texture. Therefore, the TIR bands of Landsat-8 could significantly improved the quality and
M
availability of TIR remote sensing data for geological mapping.
d
AIG approach has been applied on OLI/TIRS Landsat-8 bands. These bands linearly transformed
te
using minimum noise fraction (MNF) methods. MNF component images show steadily decreasing image quality with increasing band number, so images with higher eigenvalues
Ac ce p
contain higher spectral information [51]. RGB color combination image was assigned to three high eigenvalues MNF transformed bands. Figure 6 shows the resultant image, lithological units and hydrothermally altered rocks are discriminated in the image. Elliptical and circular patterns of hydrothermally altered rocks are appeared as purple color around known copper deposits in the study area (Fig. 6). However, the disturbances of sedimentary rocks that appeared as light purple to red color can be seen in the northern and southern parts of the image. Sedimentary rocks such as mudstone, shale, claystone and litharenite sandstones contain large amounts of detrital clays such as montmorillonite, illite and kaolinite. Detrital clays in sedimentary rocks can be erroneously mapped as hydrothermal alteration clay minerals. It should be noted that
13 Page 13 of 32
hydrothermally altered rocks associated with porphyry copper mineralization are located in crystalline igneous rocks background. The existence of thermal infrared spectral information in
which is manifested as light blue color in the scene (Fig. 6).
ip t
the MNF transformed images assist to identification of crystalline igneous rocks background,
The output of MTMF method is a set of rule images given as MF and infeasibility scores for
cr
each pixel related to end-members. Figure 9 shows MF score image of clay and carbonate
us
minerals in short-wave infrared bands of Landsat-8 for the study area. Green areas are high digital number (DN) value pixels above the background with low infeasibility (Fig. 7). Most of
an
the identified areas are associated with the known porphyry copper deposits and few of them can be seen in the sedimentary rock background (Northwestern part of the scene) in the study area.
M
The spatial distribution of the identified hydrothermally altered rocks has been verified through
d
in situ inspection. Geological locations recorded by a Garmin® eTrex Legend®H GPS. Samples
te
for XRD analysis and spectral reflectance measurements were collected from Sar Cheshmeh mines and surrounding areas. The field photographs of the geomorphology, rock units,
Ac ce p
hydrothermally altered rocks, sampling sites and open-pit quarry of copper mines are collected. According to X-ray diffraction (XRD) analysis of collected rock samples, the minerals predominantly was detected in altered rocks included muscovite, illite, kaolinite, epidote, chlorite, calcite and quartz. The average spectra measurements of collected rock samples from hydrothermally altered rocks are shown in Figure 8. The reflectance of samples from open-pit quarry of the Sar Cheshmeh mine exhibited strong absorption features at 2200 and 2170 nm. Samples from surrounding areas showed strong absorption features at 2350 nm. Fifty selected points of alteration mapped pixel points derived of from the image possessing results and real points on the ground obtained by GPS survey showed RMSE= 0.7865 for this study. It is shown
14 Page 14 of 32
that the Landsat-8 data can yield significant geological information to identify hydrothermally altered rocks and lithological units associated with porphyry copper deposits in arid and semi-
ip t
arid regions.
cr
5. Conclusions
us
This study evaluates the applicability of Landsat-8 data to extract geologic information for
an
hydrothermal alteration and lithological mapping associated with porphyry copper deposits using some selected image processing methods. Sar Cheshmeh copper mining districts in southeastern
M
segment of the Urumieh-Dokhtar Volcanic Belt, SE Iran has been selected as case study. Results indicate that Landsat-8 bands have great ability to yield spectral information for identifying
d
vegetation, iron oxide/hydroxide and clay and carbonate minerals, silicate mineral and
te
lithological units for porphyry copper exploration purposes. The TIR bands of Landsat-8
Ac ce p
significantly improved the quality and availability of TIR remote sensing data for lithological mapping. This investigation demonstrates significant implications for geologists to utilize Landsat-8 OLI/TIRS data for porphyry copper deposits exploration in the future.
Acknowledgements
This study was conducted as a part of Fundamental Research Grant scheme (FRGS) (Vote no: R.J130000.7809.4F455) granted by Universiti Teknologi Malaysia (UTM). We are thankful to the Universiti Teknologi Malaysia for providing the facilities for this investigation. We also would like to express our great appreciation to Prof. Dr. Moustafa Y El-Nagger and the anonymous reviewer for their very useful and constructive comments and suggestions for improvement of this manuscript. 15 Page 15 of 32
References
ip t
[1] L.C. Rowan, Wetlaufer, P.H., Goetz, A.F.H., Billingsley, F.C., Stewart, J.H. Discrimination
cr
of rock types and detection of hydrothermally altered areas in south-central Nevada. U.S. Geological Survey Professional Paper, 883, (1974) 35.
us
[2] A.F.H. Goetz, Rock, B.N., Rowan, L.C. Remote sensing for exploration: An overview. Economic Geology, 78, (1983) 573–590.
an
[3] M.J. Abrams., Brown, D., Lepley, L., Sadowski, R. Remote sensing of porphyry copper deposits in Southern Arizona. Economic Geology 78, (1983) 591-604.
M
[4] L.C. Rowan, Goetz, A.F.H., Ashley, R.P. Discrimination of hydrothermally altered and
522–535.
d
unaltered rocks in visible and near infrared multispectral images. Geophysics, 42 (3), (1977)
te
[5] J.K. Crowley, Brickey, D.W., Rowan, L.C. Airborne imaging spectrometer data of the Ruby
Ac ce p
Mountains, Montana: mineral discrimination using relative absorption band-depth images. Remote Sensing of Environment, 29(2), (1989) 121-134. [6] T. Cocks, Jenssen, R., Stewart, A., Wilson, I., Shields, T. The HyMap Airborne Hyperspectral Sensor: The System, Calibration and Performance. Proc. 1st EARSeL Workshop on Imaging Spectroscopy (M. Schaepman, D. Schläpfer, and K.I. Itten, Eds.), (68 October 1998), Zurich, EARSeL, Paris, p. 37-43. [7] Kruse, F.A., Boardman, J.W., Lefkoff, A.B., et al. The Spectral Image Processing System (SIPS) – Interactive Visualization and Analysis of Imaging Spectrometer Data. Remote Sensing of Environment. 44, (1993) 145-163.
16 Page 16 of 32
[8] A.F.H. Goetz, A.F.H. Three decades of hyperspectral remote sensing of the Earth: A personal view. Remote Sensing of Environment, (2009)113, S6-S16.
ip t
[9] B. A. Pour, Hashim, M., Marghany, M. Using spectral mapping techniques on short wave infrared bands of ASTER remote sensing data for alteration mineral mapping in SE Iran.
cr
Inter.J. Physical Scien, 6(4), (2011) 917-929.
[10] B.A. Pour, Hashim, M. Identification of hydrothermal alteration minerals for exploring of
us
porphyry copper deposit using ASTER data, SE Iran. Journal Asian Earth Sciences 42,
an
(2011a) 1309-1323.
[11] B.A. Pour, Hashim, M. Spectral transformation of ASTER and the discrimination of
M
hydrothermal alteration minerals in a semi-arid region, SE Iran. International Journal Physical Sciences 6(8), (2011b) 2037-2059.
d
[12] B.A Pour, Hashim M. The Earth Observing-1 (EO-1) satellite data for geological mapping,
te
southeastern segment of the Central Iranian Volcanic Belt, Iran. International Journal
Ac ce p
Physical Sciences 6(33) (2011C) 7638-7650. [13] B.A. Pour, Hashim, M. The application of ASTER remote sensing data to porphyry copper and epithermal gold deposits. Ore Geology Reviews (2012a). 44, 1-9. [14] B.A. Pour, Hashim, M. Identifying areas of high economic-potential copper mineralization using ASTER data in Urumieh-Dokhtar Volcanic Belt, Iran. Advances in Space Research 49, (2012b) 753-769.
[15] B.A Pour, Hashim M (2013). Fusing ASTER, ALI and Hyperion data for enhanced mineral mapping. International Journal of Image and Data Fusion, 4, 126-145.
17 Page 17 of 32
[16] B.A Pour, Hashim, M, and van Genderen. J. Detection of hydrothermal alteration zones in a tropical region using satellite remote sensing data: Bau gold field, Sarawak, Malaysia. Ore
ip t
Geology Reviews. 54, (2013) 181-196. [17] B.A. Pour, Hashim, M, Marghany. M Exploration of gold mineralization in a tropical region
cr
using Earth Observing-1 (EO1) and JERS-1 SAR data: a case study from Bau gold field,
us
Sarawak, Malaysia. Arabian Journal of Geosciences, 7, (2014) 2393-2406.
[18] B.A. Pour, Hashim, M. Alteration mineral mapping using ETM+ and hyperion remote
an
sensing data at Bau Gold Field, Sarawak, Malaysia. IOP Conference Series: Earth and Environmental Science 18 (1), (2014a) 012149.
M
[19] B.A. Pour, Hashim, M., 2014b. Structural geology mapping using PALSAR data in the Bau
d
gold mining district, Sarawak, Malaysia. Advances in Space Research, 54(4), 644-654.
te
[20] Pour, B. A., Hashim, M., 2014c. ASTER, ALI and Hyperion sensors data for lithological
Ac ce p
mapping and ore mineral exploration. Springerplus, 3(130), 1-19. [21] B.A. Pour, Hashim, M., 2015. Structural mapping using PALSAR data in the Central Gold Belt, Peninsular Malaysia. Ore Geology Reviews, 64, 13-22. [22] Sabins, F.F. Remote sensing for mineral exploration. Ore Geology Reviews 14, ( 1999) 157183.
[24] National Aeronautics and Space Administration. Earth Observing-1 Advanced Land Imager. Online: http://eo1.gsfc.nasa.gov/Technology/ALIhome1.htm (2002). [25] National Aeronautics and Space Administration. Earth Observing-1 EO1General Mission. Online: http://eo1.gsfc.nasa.gov/new/general/ (2004).
18 Page 18 of 32
[26] Wulder, M.A., White, J.C., Goward, S.N., Jeffrey G.M., Irons, J.R., Herold, M., Cohen, W.B., Loveland, T.R., Woodcock, C.E. Landsat continuity: Issues and opportunities for land cover monitoring. Remote Sensing of Environment 112, (2008) 955-969.
ip t
[27] D.R. Hearn, Digenis, C.J., Lencioni, D.E., Mendenhall, J.A., Evans, J.B., Walesh, R.D. EO-
and Remote Sensing (2001) p: 897-899.
us
[28] USGU (2012) Landsat Data Continuity Mission. July 2012.
cr
1 Avanced Land Imager overview and spatial performance. IEEE Transaction. Geoscience
an
[29] H. Hubner, Geological map of Iran sheet no. 5, southcentral Iran: Tehran, National Iranian Oil Company, scale 1:1,000,000. 1969a.
M
[30] J. C. Mars, Rowan, L.C. Regional mapping of phyllic- and argillic-altered rocks in the Zagros magmatic arc, Iran, using Advanced Spaceborne Thermal Emission and Reflection
te
186.
d
Radiometer (ASTER) data and logical operator algorithms. Geosphere 2(3), (2006) 161-
Ac ce p
[31] Waterman, G.C., Hamilton, R.L. The Sarcheshmeh porphyry copper deposit. Economic Geology 70, (1975) 568-576.
[32] H. Atapour, and Aftabi, A. The geochemistry of gossans associated with Sarcheshmeh porphyry copper deposit, Rafsanjan, Kerman, Iran: Implications for exploration and the environment. Journal of Geochemical Exploration, 93, (2007) 47-65. [33] A. Hezarkhani. Hydrothermal evolution of the Sar-Cheshmeh porphyry Cu–Mo deposit, Iran: Evidence from fluid inclusions. Journal of Asian Earth Sciences, 28, (2006) 409-422. [34] Singer, D.A., Berger, V.I. and Moring, B.C. Porphyry copper deposits of the world: Database and grade and tonnage models: U.S. Geological Survey Open-File Report 2008−1155 (http://pubs.usgs.gov/of/ 2008/1155). 19 Page 19 of 32
[35] H. Ranjbar, M. Honarmandb, Z. Moezifar. Application of the Crosta technique for porphyry copper alteration mapping, using ETM+ data in the southern part of the Iranian volcanic sedimentary belt. J. Asian Earth Sci. 24, 237–243. 2004.
ip t
[36] H. Ranjbar, Masoumi, F., Carranza, E.J.M. Evaluation of geophysics and spaceborne multispectral data for alteration mapping in the Sar Cheshmeh mining area, Iran.
cr
International Journal of Remote Sensing, 32 (12), (2011) 3309-3327.
us
[37] M. Hosienjani Zadeh., Tangestani, M. H., Roldan, F. V., & Yusta, I. Sub-pixel mineral mapping of a porphyry copper belt using EO-1 Hyperion data. Advances in Space Research
an
53(3) (2013), 440-451.
[38] M. Honarmand, Ranjbar, H., and Shahabpour, J. Combined use of ASTER and ALI data for
M
hydrothermal alteration mapping in the northwestern part of the Kerman magmatic arc, Iran.
d
International Journal of Remote Sensing, 34(6), (2013) 2023-2046.
te
[39] E. Ben-Dor, Kruse, F.A., Lefkoff, A.B. and Banin, A. Comparison of three calibration techniques for utilization of GER 63-channel aircraft scanner data of Makhtesh Ramon,
Ac ce p
Nega, Israel. Photogrammetric Engineering and Remote Sensing, 60(11), (1994) 1339-1354. [40] Scheidt, S., Ramsey, M. and Lancaster, N. Radiometric normalization and image mosaic generation of ASTER thermal infrared data: An application to extensive sand and dune fields. Remote Sensing of Environment, 112(3), (2008) 920−933. [41] G.R. Hunt. Spectral signatures of particulate minerals in the visible and near infrared. Geophysics. 42, (1977) 501-513. [42] G.R. Hunt, Ashley, P. spectra of altered rocks in the visible and near infraraed. Economic Geology. 74, (1979) 1613-1629.
20 Page 20 of 32
[43] G.R. Hunt and Salisbury, J. W. Mid-infrared spectral behavior of metamorphic rocks, Technical Report AFRCL-TR-76-0003, US Air Force Cambridge Research Laboratory, Cambridge, MA (1976).
ip t
[44] I. Di Tommaso, and Rubinstein, N. Hydrothermal alteration mapping using ASTER data in the Infiernillo porphyry deposit, Argentina. Ore Geology Reviews, 32, (2007) 275-290.
cr
[45] J. Inzana, Kusky, T., Higgs, G ., Tucker, R. Supervised classifications of Landsat TM band
us
ratio images and Landsat TM band ratio image with radar for geological interpretations of central Madagascar. Journal of African Earth Sciences 37, (2003) 59-72.
an
[46] J.W. Boardman., F.A. Kruse, and R.O. Green. Mapping target signatures via partial unmixing of AVIRIS data, Summaries, Proceedings of the Fifth JPL Airborne Earth Science
M
Workshop, 23–26 January, Pasadena, California, JPL Publication 95–1, v. 1,( 1995) pp. 23–
d
26.
te
[47] J. W. Boardman. Leveraging the high dimensionality of AVIRIS data for improved subpixel target unmixing and rejection of false positives: mixture tuned matched filtering, in:
Ac ce p
Summaries of the Seventh Annual JPL Airborne Geoscience Workshop, Pasadena, CA, (1998) pp. 55.
[48] A.A. Green, Berman, M., Switzer, P., Craig, M. D. A transformation for ordering multispectral data in terms of image quality with implications for noise removal: IEEE Transaction. Geoscience and Remote Sensing 26(1),( 1988) 65-74. [49] Ruiz-Armenta, J. R. and Prol-Ledesma, R. M. Techniques for enhancing the spectral response of hydrothermal alteration minerals in Thematic Mapper images of Central Mexico. International Journal of Remote Sensing 19, (1998) 1981-2000.
21 Page 21 of 32
[50] Watanabe, H., Matsuo, K., Rock type classification by multi-band TIR of ASTER. Geosciences Journal 7, (2003) 347–358. [51] C.M. Chen. Comparison of principal component analysis and Minimum Noise Fraction
ip t
transformation for reducing the dimensionality of hyperspectral imagery. Geography
us
cr
Research. 33, (2000)163-178.
Figure Captions
an
Fig.1. Simplified geology map of southeastern segment of the Urumieh–Dokhtar volcanic Belt
M
(modified after Pour and Hashim, 2012b). Study areas are located in rectangles. Fig.2. Geological map of Sar Cheshmeh region (modified after Mars and Rowan, 2006).
d
Fig. 3. (A) Laboratory spectra of epidote, calcite, muscovite, kaolinite, chlorite and alunite. (B)
te
Laboratory spectra of limonite, jarosite, hematite and goethite (Clark et al., 1993).
Ac ce p
Fig.4. Band ratios of 4/2, 6/7, 5 in RGB.
Fig. 5. Band ratios of 4/2, 6/7, 10 in RGB. Fig. 6. RGB color combination of MNF 1, 2, 3 components. Fig.7. MF score image of clay and carbonate minerals. Fig.8. Laboratory reflectance spectra of altered rock samples from alteration zones in the study area.
Table captions 22 Page 22 of 32
Table 1. The performance characteristics of the ALI and ETM+ and Landsat-8 OLI/TIRS
Ac ce p
te
d
M
an
us
cr
ip t
sensors (Bryant et al., 2003; Beck, 2003; Lobell and Asner, 2003).
23 Page 23 of 32
Table 1. The performance characteristics of the ALI and ETM+ and Landsat-8 sensors.
0.480-0.690 0.433-0.453 0.450-0.515 0.525-0.605 0.633-0.690 0.775-0.805 0.845-0.890 1.200-1.300 1.550-1.750 2.080-2.350
Pan 1 2 3 4 5 7 6
0.520-0.900 0.450-0.515 0.525-0.605 0.633-0.690 0.780-0.900 1.550-1.750 2.090-2.350 10.45-12.50
1 2 3 4 5 6 7 Pan 9 10 11
0.433-0.453 0.450-0.515 0.525-0.600 0.630-0.680 0.845-0.885 1.560-1.660 2.100-2.300 0.500-0.680 1.360-1.390 10.30-11.30 11.50-12.50
VNIR
SWIR TIR Landsat-8 VNIR
TIR
37
185
30
185
15
Ac ce p
SWIR
14.25 28.50
te
ETM+
10 30
ip t
Pan 1 2 3 4 5 6 7 8 9
Swath Width(km)
cr
VNIR
Ground Resolution(m)
us
ALI
SWIR
Spectral Range(μm)
an
Band Number
M
Subsystem
d
Sensors
100
The visible near-infrared (VNIR), the short-wave infrared (SWIR) and the thermal-infrared (TIR) are
abbreviated.
24 Page 24 of 32
Ac
ce
pt
ed
M
an
us
cr
i
Figure 1
Page 25 of 32
Ac
ce
pt
ed
M
an
us
cr
i
Figure 2
Page 26 of 32
Ac
ce
pt
ed
M
an
us
cr
i
Figure 3
Page 27 of 32
Ac ce p
te
d
M
an
us
cr
ip t
Figure 4
Page 28 of 32
Ac ce p
te
d
M
an
us
cr
ip t
Figure 5
Page 29 of 32
Ac ce p
te
d
M
an
us
cr
ip t
Figure 6
Page 30 of 32
Ac ce p
te
d
M
an
us
cr
ip t
Figure 7
Page 31 of 32
Ac
ce
pt
ed
M
an
us
cr
i
Figure 8
Page 32 of 32
