Download 5. Results and discussion

2014년 인공위성 원격탐사
ASTER detection of chromite bearing
mineralized zones in Semail Ophiolite Massifs
of the northern Oman Mountains: Exploration strategy
(SCI, I.F 2.417, 2012)
Sankaran Rajendran
Department of Earth Sciences, Sultan Qaboos University, Al-Khod, 123 Muscat, Oman
2014. 5.
에너지∙자원공학과
손영진
Contents
1. Introduction
2. Satellite data
3. Spectral characteristics of ophiolites and chromite
4. Methodology
5. Results and discussion
Chromites 크롬철광(크로마이트)
Harzburgite 감람암(하아즈버어자이트)
Ophiolite 사문석(오피오라이트)
Remote sensing 원격탐사
Semail massifs 세마일 대산괴(단층 지괴)
ultramafic rocks 초염기성암
Serpentinized 사문암화
Gabbro 반려암
Abstract
 Chromite deposit
in the ultramafic rocks of Semail ophiolite massifs of the northern Oman Mountains
 Geological mapping using conventional mapping techniques
difficult access, complexity of structures, lack of resolution
 The present research study
Discrimination and occurrence of chromites bearing mineralized zones within ophiolites
 Analysis method
Landsat TM & ASTER satellite data
→ decorrelated stretching, different band rationing, Principal Component Analysis(PCA)
 Results
Processed VNIR, SWIR spectral wavelength regions are promising
be successful for mapping of serpentinized harzburgite containing chromites
1. Introduction
 Geology
→ The Semail ophiolite of Oman (Fig. 1a) is one of the world's best and most
complete examples of well-exposed ocean floor rocks (Abrams et al., 1988)
→ It mainly consists of two main lithological units:
(1) serpentinized mantle ultramafic rocks including harzburgite with dunite
(2) gabbroic–basaltic rocks from the overlying crustal section
→ The hot base of the Semail ophiolite developed a sole of metamorphic rocks
formed at high temperature (T) and low pressure (P) during interoceanic detachment
(Bucher et al., 1988; Ghent and Stout, 1981; Searle and Malpas, 1980)
→ Chromite have been found recently deeper within the mantle section of the Semail
opholite massifs
→ Discrimination of mantle ultramafic harzburgite with dunite(partially, serpentinized)
Delineation of the areas of potential chromite mineralized zones within the Semail
ophiolite massifs in this region
1. Introduction
 Conventional method (detailed mapping
are impossible, study area)
2
→ the extent of the area (1169 km )
→ extremely rugged topography
→ elevations between 0 and 2500 m
→ exhaustive sampling
 Remote sensing
→ Remote sensing is able to provide more detailed mapping of the areas of chromite
mineralized zones in the ultramafic section of the ophiolite
→ It facilitates the exploration geologists, industrialists and mine owners for cost saving
exploitation
→ Remote sensing is most suitable technique to apply in arid regions like Oman
where there are rock exposures over the surface and no vegetation disturbance
1. Introduction
 Purpose
→ The discrimination and occurrence of chromites bearing mineralized zone
by analyzing the capabilities of Landsat TM and ASTER satellite data
 Image processing techniques
→ the decorrelated stretching
→ different band rationing
→ Principal Component Analysis (PCA)
 Results
→ Detecting the areas of potential chromite bearing mineralized zones
within the ophiolite region
→ Mapping of serpentinized harzburgite containing chromites
→ Exploration geologists, industrialists andmine owners are advised
to adopt this technique and avoid the limits in field data in arid region elsewhere
1. Introduction
Fig. 1
2. Satellite data
 To discriminate the harzburgites and the area of chromite
A comparative study is performed between Landsat TM and ASTER spectral bands
 Landsat TM data: on March 4, 2003 with 0% cloud cover
 ASTER sensor data: Earth Observing System (EOS) TERRA platform
December 1999, inclination of 98.2°, altitude of 705 km, a repeat cycle of 16 days
14-band ASTER Level 1B data acquired on April 18, 2006
2. Satellite data –
Landsat TM
Landsat TM(참고자료)
 History
→ 지구관측을 위한 최초의 민간목적 원격탐사 위성
→ 1972년에 1호 위성이 발사
→ LANDSAT 2,3,4,5 호가 차례로 발사에 성공
→ LANDSAT 6는 궤도 진입에 실패
 Operation
→ LANDSAT 7호가 1999년 4월에 발사
→ 현재 1,2,3,4호는 임무를 끝내고 운영이 중단
→ 현재는 5,7호만 운용 중
→ Thematic Mapper(TM), Multispectral Scanner(MSS)를 탑재
2. Satellite data –
Landsat TM
Landsat TM(참고자료)
Landsat 위성자료를 무료로 다운로드 링크
http://earthexplorer.usgs.gov/
2. Satellite data –
ASTER
ASTER(참고자료)
→ Advanced Spaceborne Thermal Emission and Reflection Radiometer
→ NASA와 일본의 METI(Ministry of Economy, Trade and Industry) 그리고
ERSDAC (Earth Remote Sensing Data Analysis Center)에 의하여
1999년 12월에 발사된 지구 탐사 위성에 탑재된 센서
→ 센서:
VNIR (Visible and Near Infrared), WIR (Shortwave Infrared), TIR(Thermal Infrared)
→ 해상도: 15m, 30m, 90m의 해상도
→ 특징 (1) 기존의 영상에 비하여 가격대비 높은 분광 해상도
→ 특징 (2) visible, near IR, short wave IR and Thermal IR의 분광영역이 넓음
→ 특징 (3) 스테레오 영상을 제공
→ 특징 (4) 육지에서의 화산 활동, 해안선의 변화, 열대우림에서의 식생,
습지등의 모니터링과 지표면 에너지 흐름에 대한 예측,
그리고 일반적인 DEM을 추출하는데 사용 가능
2. Satellite data –
ASTER(참고자료)
ASTER
3. Spectral characteristics of ophiolites and chromite
 Use of remote sensing
→ To implement lithological mapping as well as identifying mineral deposits
(Abrams and Hook, 1995; Abrams et al., 1983; Gad and Kusky, 2007; Gabr et al., 2010; Kusky and Ramadan,
2002; Mars and Rowan, 2006; Rowan and Mars, 2003; Rowan et al., 2003; Sabins, 1997)
→ For mapping of hydrothermally altered minerals that have distinct absorption
features(Hunt,1977)
→ To explore mineral in at least four ways
(Sabins, 1997)
1) mapping regional lineaments and structural trends along which groups of mining
may occur
2) mapping local fracture patterns that may control individual ore deposits
3) detecting hydrothermally altered rocks associated with ore deposits
4) providing basic geologic data
3. Spectral characteristics of ophiolites and chromite
Absorption
 Increase of absorption in bands in the visible and short wavelength infrared
→ Due to either electronic or vibrational processes that occur in these minerals
 Earlier spectral reflectance studies provide important insights to the causes of
spectral variations for the interpretation of optical remote sensing data
(e.g., Cloutis and Gaffey, 1991; Cloutis et al., 2004)
 Abrams et al. (1988) presented a detailed review about ophiolite lithology derived
from Hunt and Salisbury (1970), Hunt et al. (1974) and Hunt (1977)
 Cr2O3 content ranging from 31.5 to 54.61 wt.%
 With Cr:Fe ratio from 1.84:1 to 2.96:1(Annells, 1989)
 And many others…
3. Spectral characteristics of ophiolites and chromite
The laboratory reflectance spectra(Abrams ea al., 1988)
 The range of spectral features produced
by some of the ophiolite minerals
 To idealize spectral reponse, collect spectra from
two unweathered ophiolite lithologies
 Serpentinite
→ A reltaively flat spectral response
3+
→ 0.45 μm is due to ferric iron (Fe )
→ Broader absorption centered near 0.9 μm
2+
or around 1.0 μm is due to ferrous iron (Fe )
→ 2.3 μm is due to vibrational processes of MG-OH
→ Hydration effects: the absorption near 1.4 μm
 Harzburgite spectrum = ferrous iron(1.0 μm)
→ attributable to ferrous iron in pyroxene and olivine
Fig. 2
3. Spectral characteristics of ophiolites and chromite
The electronic processes in the minerals
 3 types: conduction bands, charge transfer and crystal field effects
2+
→ ferrous iron (Fe ) in weathered surface produces absorption centered
at about 0.45 μm, 1.0–1.1 μm, 1.8–1.9 μm, and 2.2–2.3 μm
3+
→ ferric iron (Fe ) - 0.65 μm and 0.87 μm
The vibrational processes which cause VNIR, SWIR absorption
→ Al–OH and Mg–OH
→ Al–OH produces absorptions centered at about 2.2 μm
→ Mg–OH produces features at about 2.3 μm
4. Methodology
Analysis methods
 Decorrelation stretching
→ Abrams et al. (1988) mapped the Oman ophiolite using Landsat TM data
by the processing of decorrelation stretching
→ Decorrelation stretching of these data produces images (RGB: 7, 5 and 4)
→ Attemption to discriminate the harzburgite and chromite using Landsat TM data
 The band ratio images
→ The band ratio images are used to suppress the topographic variation
and the image brightness difference related to grain size variation
(Adam and Felic, 1967; Sultan et al., 1987)
→ All the reasonable grouping of minerals of such rocks is best discriminated
by a combination of ratios
→ RGB band ratios used to discriminate serpentinites by Sultan et al. (1986) (5/7, 5/1, 5/4×4/3),
Sabins (1999) (3/5, 3/1, 5/7) and Gad and Kusky(2006) ((5/3, 5/1, 7/5) and (7/5, 5/ 4, 3/1))
 PCA technique
→ A PCA is applied to the 9 ASTER bands of study area (e.g., Richards and Xiuping, 1998)
→ PCs contain the spectral information due to minerals and whether the digital
numbers (DNs) of pixels containing the target minerals high (bright) or low (dark)
values (Crósta and Filho, 2003; Crósta and Moore, 1989; Loughlin, 1991; Rokos et al., 2000)
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Landsat TM mapping(Decorrelation stretching)
 The decorrelated image processed
using the three infrared bands (TM bands 7, 5 and 4)
displayed in red, green and blue(RGB)
 Band 4 information relating to the presence of iron
 Band 5 characterizes the general albedo of
the materials to highlight certain altered
(ferric iron-bearing) rocks
 Band 7 responds to the presence of hydroxyl-bearing
minerals
E — Basic extrusives mostly spilites with pillow lava or conglomerate
D — Diabase dyke swarms
G— Gabbro
HG—Gabbroid hypabyssal rocks
PG — Cumulate layered gabbro
P and CD — Sheared serpentinized harzburgite
evidence on the occurrences of chromites(star marked)
Fig. 3
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Landsat TM mapping(Decorrelation stretching)
Fig. 4
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Landsat TM mapping(The band ratio images)
 Landsat RGB band ratio image
→ Sultan et al. (1986) (5/7,5/1, 5/4*4/3)
→ Sabins (1999) (3/5, 3/1, 5/7)
→ Gad and Kusky (2006) ((5/3, 5/1, 7/5) and (7/5, 5/4, 3/1))
 These can be studied with the decorrelated
RGB image (Fig. 3) and geology (Fig. 1c)
 a,b(Sultan et al., Sabins)
→ the serpentinized harzburgite in dark green-yellow color
→ owing to the absorption by MgO- and OH-bearing minerals
→ high degree of absorption of iron bearing minerals
 c,d(Gad and Kusky)
→ the serpentinized harzburgites in a dark violet-bluish color
→ the area of chromite mineralized zone
in dark purple (3/5, 3/1, 5/7 and 5/3,5/1, 7/5)
and pink (7/5, 5/4, 3/1) colors
Fig. 5
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 USGS spectral library by Amer et al. (2010)
→ Serpentine(main component of serpentinized harzburgites) has two absorption
features at 1.4 and 2.35 μm
→ Gabbros and metabasalts consist mainly of pyroxene with an absorption feature
at 2.35 μm
→ Anorthitic plagioclase has two absorption features around 2 and 2.4 μm
→ Olivine has absorption features between 0.8 and 1.2 μm
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 ASTER band ratios (4/7, 4/1, 2/3*4/3; Fig. 6a) of Abdeen et al. (2001) are equivalent
to the Landsat TM image (5/7, 5/1, 3/4*5/4; Fig. 5a) of Sultan et al. (1986)
 ASTER band ratio image (4/7, 3/4, 2/1; Fig. 6b) of Abdeen et al. (2001) is equivalent
to the Landsat TM image of Abram's combination (5/7, 4/5, 3/1)
 The image processed on band ratio (4/7, 4/1, 2/3*4/3) and (4/7, 3/4, 2/1) of study
area shows the serpentinized harzburgites in brown-yellowish and green-pinkish
colors

The area mineralized zone(well discriminated)
red-pinkish color by the band ratio(4/7, 4/1, 2/3*4/3)
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 Current image spectra(Amer et al., 2010)
 A new ASTER band ratio ((2+4)/3, (5+7)/6, (7+9)/8)
 (The numerator is the sum of the bands representing the shoulders)
 (The denominator is the band located nearest the absorption feature minimum)
 The ratio is best for lithological mapping in arid regions
→ the band ratio (2+4)/3 highlights the serpentinites and metabasalts
→ (5+7)/6 highlights the granitic rocks
→ (7+9)/8 highlights the metagabbros and serpentinites
 These band ratios were used by them in red, green and blue to produce a color
image to distinguish between serpentinites from the other ophiolitic rocks
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 The geologic interpretations of ASTER band
ratios ((2+4)/3, (5+7)/6, (7+9)/8) image (Fig. 6c)
→ The sharp contact of serpentinized harzburgites
between other ophiolitic rocks well
Fig. 6
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 Principal Component Analysis of ASTER bands
→ the image of the first five components (Fig. 7)
→ the PCA 2nd component (Fig. 7b) highlights the
serpentinized harzburgites
→ the PCA 4 (Fig. 7d) highlights the basic gabbros
→ the PCA 5 component (Fig. 7e) shows some
interesting signatures on the post tectonic
younger metabasalts, the extrusives
 The selection of three bands(PC5, PC4 and PC2)
(Amer et al., 2010)
→ For better discrimination of serpentinized harzburgites
→ For delineation of the area of potential chromite mineralized zone
Fig. 7
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
ASTER mapping
 The RGB results of PC5, PC4 and PC2(Fig. 8.)
 Discrimination
of highly sheared serpentinized harzburgites
→ By pink color (CD: upper part of sequence)
→ By greenish yellow (P: lower part of sequence)
 Delineation
of potential chromites bearing mineralized zone
→ from pink and greenish yellow colors
within the serpentinized harzburgites
 The PCA is powerful in distinguishing the subtle
differences between the various rock units of the study area.
Fig. 8
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Field interpretation
 To evaluate the VNIR and SWIR regions of Landsat TM and ASTER satellite data
→ For discrimination ophiolitic rocks and the area of chromite bearing mineralized
zones
 Field-checks on Semail ophiolite massifs(during November–December 2010)
→ With existing geological maps and interpreted satellite images
 Real lithological information and detection of area of chromite
bearing mineralized zones
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Field interpretation

In the field
Semail Ophiolite
Massifs
→ The Semail ophiolite massifs of study
area (Fig. 9a) covers the most complete
ophiolite succession
→ Comprises highly sheared
serpentinized harzburgites (Fig. 9b)
Fig. 9
Serpentinized
Harzburgite
Chromite vein
Layered Gabbro
→ Layered gabbro (Fig. 9c)
→ Diabase dykes (Fig. 9d)
Serpentinized
Harzburgite
Chromite vein
Serpentinized
Harzburgite
Chromite vein
5. Results and discussion
Discrimination of ophiolites and delineation of chromite mineralized zone
Field interpretation
Fig. 4
Fig. 10
6. Conclusions
 The Landsat TM and ASTER satellite data processed in VNIR and SWIR spectral
wavelength regions
 To discriminate of ophiolite lithologies (harzburgite, metagabbro, and metabasalt)
 To delineate of the area of chromite mineralized zone within ophiolites
 Methods: Decorrelation stretching, Band ratios, principal component analysis(PCA)
 Comparison of processed images with `Google Earth` image
 Validation through field work
 Occurrence of chromite mineralized zone below the Moho run about 1 to 5 km
within mantle sequences
 The discussed techniques have potential, as time- and cost-effective methods, for
mineral exploration targeting additional chromite deposits in comparison to the time
consuming classical field mapping and exploration, especially in inaccessible areas.