Earth Sci Inform
DOI 10.1007/s12145-015-0239-x
RESEARCH ARTICLE
Reflectance spectroscopy and remote sensing data for finding
sulfide-bearing alteration zones and mapping geology
in Gilgit-Baltistan, Pakistan
Laeiq Ahmad 1 & M. Tahir Shah 1 & Shuhab D. Khan 2
Received: 28 November 2014 / Accepted: 1 September 2015
# Springer-Verlag Berlin Heidelberg 2015
Abstract Gilgit-Baltistan region is covering the northern
most part of Pakistan where the rocks of the Kohistan-
Ladakh island arc and Karakoram plate are exposed. The area
has greater potential for precious and base metals deposits
which are needed to be explored through spectroscopy and
remote sensing techniques. Minerals and rocks can nowadays
be identified through the measurement of their absorption and
reflectance features by spectroscopic analysis. Spectral reflec-
tance analysis is also very important in selecting the appropri-
ate spectral bands for remote-sensing data analysis of un-
known or inaccessible areas. In this study, reflectance spectra
in the spectral range of 0.35–2.5 μm of different types of
unaltered and altered rocks found in the Machulu and Astor
areas of northern Pakistan were obtained using an ASD
spectroradiometer. The fresh rock samples showed low spec-
tral reflectance as compared to the altered rock samples. The
minerals jarosite, goethite, and hematite showed depth of ab-
sorption minima in the range of 0.4–1.15 μm due to the pres-
ence of iron (Fe), while jarosite and limonite showed absorp-
tion depth at 2.2 μm due to the presence of hydroxyl ions
(OH¯). The clay minerals montmorillonite and illite showed
absorption depth at 1.93 and 2.1 μm, respectively. Muscovite
showed depth of absorption minima at 1.4 and 1.9 μm in some
samples. Calcite showed deep absorption minima at 2.32 μm,
while anorthite showed absorption features at 1.4, 1.9, 2.24,
Communicated by: H. A. Babaie
* Laeiq Ahmad
laeiqgeo@yahoo.com
1
National Centre of Excellence in Geology, University of Peshawar,
Peshawar 25120, Pakistan
2
Department of Earth and Atmospheric Sciences, University of
Houston, Houston, TX, USA
and 2.33 μm. Olivine showed a slight depressed absorption
feature at 1.07 μm. The copper-bearing phases malachite,
chrysocolla, and azurite showed, respectively, a broad absorp-
tion feature in the range of 0.6–0.9 μm, a small absorption at
1.4 μm, and a deep absorption at 1.93 μm. The unmineralized
samples exhibited high reflectance in the wavelength ranges
of 0.6–0.8, 1.6–1.9, 2.0–2.3, 2.1–2.25, and 2.4–2.5 μm, re-
spectively, while the mineralized samples showed reflectance
bands in the wavelength ranges of 0.4–0.6, 1.3–1.8, and 2.1–
2.2 μm. On this basis, the band ratio combinations 7/5–4/3–6/
3 and 7/5–6/3–4/3 of Landsat 8 and 4/7–4/3–2/1 for ASTER
data were found to be very effective in the lithological differ-
entiation of major rock units.
Keywords Reflectance spectroscopy . Landsat . ASTER .
Alteration zones . Geology . Northern Pakistan
Introduction
Gilgit-Baltistan is the northern most province of Pakistan
which is mainly accessed from Islamabad the capital city,
through Karakoram Highway (KKH). Geographically, this re-
gion is bounded by Afghanistan in the northwest, China in the
northeast and province of Khyber Pakhtunkhwa in the south-
west. This region is hosting many highest mountain peaks
with few of these are higher than 8000 m. The geology of
the Gilgit-Baltistan region is very complex in the context of
tectonic setup. This region is mainly composed of the mafic-
ultramafic complexes and batholith plutons of the Kohistan-
Ladakh island arc (KLIA) and the meta-sedimentary and
meta-igneous rocks of the Karakoram plate (Fig. 1). The
KLIA is an intra-oceanic Cretaceous island arc which is
formed as a result of the collision and subduction of the Indian
plate underneath the Asian plate along Indus Suture Zone

Fig. 1 Map showing samples location of study area Machulu and Astor
(ISZ) in the south but the timing of the collision is still con-
troversial (Molnar and Tapponnier 1975; Treloar et al. 1996;
Hanson 1989; Beck et al. 1995; Rowley 1996; Clift et al.
2002; Bignold and Treloar 2003; Tahirkheli 1979; Yin and
Harrison 2000; Ziabrev et al. 2004; Yin 2006; Khan et al.
2009; Chatterjee et al. 2013). The KLIA has northern thrusted
contact with Karakoram plate along the Northern Suture Zone
(NSZ). Detailed geological work has been done on the main
lithological units of the Gilgit-Baltistan region in regard to
their mineralogy, geochemistry and genesis. However, there
is a great need to explore the economic potential of the region
through advanced level geophysical, geochemical and spec-
troscopic and remote sensing techniques. The occurrence of
gold particles in the Indus river and its tributaries flowing
through this region (Tahirkheli 1974; Austromineral 1976,
1978; Halfpenny and Mazzucchelli 1999; Shah and Khan
2004; Ali et al. 2014; Biber et al. 2014) and the hydrothermal-
ly altered sulfide-bearing zones reported in various parts of the
region (Sheikh et al. 2014; Miandad et al. 2014; Rehman et al.
2015) suggest that there could be the possibility of occurrence
of source rocks for precious and base metals mineralization in
the region. But due to the remoteness and inaccessibility to
rugged terrains in this part of the Himalayan region, we need
Earth Sci Inform
to adopt such techniques which can scan the whole region for
the differentiation of various lithological units and hydrother-
mally altered mineralized and un-mineralized zones. During
this study the spectroscopic and remote sensing data of select-
ed areas have been used to accomplish the above mentioned
goals.
Spectroscopy is the study of light as a function of wave-
length that is emitted, reflected, or scattered from a liquid,
solid, or gas (Clark 1999). Spectral reflectance plays a vital
role in the field of remote sensing. Each element has a specific
structure, each mineral comprises a specific collection of ele-
ments, and the aggregate of these minerals is called rock. The
differences in the spectral reflectance properties of minerals
and rocks are related to their grain size, texture, and color.
Whether the material is solid, liquid, or gas, the emittance
and reflectance spectroscopies of natural surfaces are sensitive
to specific chemical bonds in the materials. Variations in the
material composition cause changes in the shape and position
of the bands in the absorption spectrum, as a result of which
each material yields a unique diagnostic vibrational band. The
conduction process, electronic transition, charge transfer, and
vibrational overtones of minerals and rocks all cause diagnos-
tic absorption bands in the visible, visible and near-infrared
Earth Sci Inform
(VNIR), and short-wave infrared (SWIR) regions of reflected
light (Clark et al. 1990). The causes of the absorption features
of minerals have been well studied by Hunt (1977), Clark
et al. (1990), Salisbury (1993) and Vincent (1997).
The ability to use multispectral remote sensing to discrim-
inate between materials depends on the differences between
their respective spectral reflectance properties. The aim of this
research was to measure the spectral reflectances of the col-
lected rocks, both unaltered and altered, from the study area to
identify anomalies in both their absorption and reflectance
features. This information was then used to identify specific
minerals and to select different bands of the electromagnetic
spectrum from the Landsat data for further interpretation.
Methodology
Rock samples were collected from unaltered and sulfide-
bearing altered rocks and quartz veins in the Machulu and
Astor areas of northern Pakistan during a field trip in the
summer of 2012 (Fig. 1). Small chips (3×2×1 cm) cut from
the rock samples were transferred to the University of Hous-
ton, Houston, USA for spectroscopic study in a controlled
laboratory environment. A portable Spectroradiometer
(Fig. 2) was used to acquire the spectral data (Fig. 2). This
spectrometer had a sampling interval of 2 to 3 μm with a full
spectral range of 0.35–2.5 μm. It also had a sufficiently fast
scan time of 100 milliseconds for a complete 0.35–2.5 μm
spectrum scan, which helped to avoid errors in collecting the
data. The High Intensity Contact Probe (Fig. 2) was used to
acquire the spectral data. The spectra for each sample were
collected by placing the probe on the flat portion of the chip. A
total of eighty-three (83) rock samples were analyzed, twenty-
Fig. 2 Portable Spectroradiometer and high intensity contact probe
which was used to collect data in the laboratory
nine (29) from the Machulu area and fifty-four (54) from the
Astor area. For each rock sample, three (3) spectra were col-
lected, and an average was calculated. The data were then
processed through ViewSpec Pro version 6.0 software,
changed to text format, and transferred to a Microsoft Excel
spreadsheet in the form of columns of wavelength and
reflectance.
Landsat 8, has nine (9) multispectral bands, including a
panchromatic band and two (2) thermal Infrared bands. The
resolution of the eight (8) visible spectral bands is 30 m, while
that of the panchromatic is 15 m, and that of the two thermal
infrared bands is 100 m. Only the VNIR, NIR, and SWIR
bands were used for the current research work. Landsat 8 level
1B data for both study areas were acquired online from United
States Geological Survey (USGS). The data were processed
using the log residual technique with ENVI (Environment for
Visualizing Images) version 5.0 software. Different false color
combinations of red-green-blue (RGB) were used. Various
band ratios were also established by using the band math
and band ratios techniques since they were found appropriate
by spectral data for mapping the geology of the study areas.
This study also utilized The Advanced Spaceborne Ther-
mal Emission and Reflection Radiometer (ASTER) data. AS-
TER data consist of fourteen bands, three of these bands are in
visible and near infrared range with spatial resolution of 15 m,
six spectral bands in SWIR are 30 m in spatial resolution.
While the remaining 5 five spectral bands cover thermal in-
frared and are 90 m in spatial resolution.
Results
Spectroscopy
This section presents spectroscopic data obtained from repre-
sentative chip samples of unaltered and sulfide-bearing altered
rocks and quartz veins found in the Machulu and Astor areas.
Machulu area
Unaltered rocks Spectroscopic analysis was carried out on
unaltered twenty-three (23) samples collected from the
Machulu area. These rock samples included fresh-looking ul-
tramafic, volcanic, dioritic, and granitic rocks. The ultramafic
rock samples showed different absorption and reflectance fea-
tures at various wavelengths. The hydrous sulfate mineral
jarosite produced a broad declining spectral absorption feature
between 0.40 and 0.50 μm. Antigorite showed absorption
features at 1.395 and 1.974 μm, while the clay mineral illite
and the carbonate calcite showed absorption at 2.1 and
2.35 μm, respectively (Fig. 3). In the volcanic rocks, positive
absorption features were noted between 0.40 and 0.50, at
1.141, at 1.399, at 1.91, and at 2.35 μm, which are

Fig. 3 Absorption and reflectance spectral features of different minerals
of ultramafic and volcanic rocks at various wavelengths of the study area
Machulu
characteristic of jarosite, phlogopite, antigorite, montmoril-
lonite, and calcite, respectively (Fig. 3). The absorption spec-
tra of the dioritic rocks showed a low declining absorption
feature at wavelength 0.946 μm, characteristic of jarosite.
Other absorption features were noted at various wavelengths,
including 1.411 and 1.914 μm, indicating anorthite,
2.247 μm, indicating calcic-plagioclase, and 2.334 μm, indi-
cating K-feldspar (Fig. 4). The spectral absorption features
shown by the granites occurred at wavelengths of 0.49 μm
for jarosite, at 1.413 and 1.92 μm for muscovite, and at
2.203 μm for muscovite and lepidolite (Fig. 4).
Sulfide-bearing altered rocks These showed more promi-
nent spectral reflectance as compared to unaltered rocks due
to the differences in mineral composition, texture, and color.
Six (6) altered rock samples, with variations in their spectral
absorption and reflectance features at different wavelengths,
were selected for spectral analysis. They were placed together
in three groups on the basis of the similarity of their spectral
features. The group-1 samples showed different peaks of re-
flectance and depths of absorption features at various wave-
lengths. Absorption features were noted at 0.434 μm, indicat-
ing jarosite, at 0.93 μm, indicating chlorite, at 1.41 μm, indi-
cating muscovite, at 1.916 μm, indicating kaolinite and smec-
tite, and at 2.203 μm, indicating muscovite and lepidolite
Fig. 4 Graph showing absorption and reflectance spectral features of
different minerals of granitic and dioritic rocks at various wavelengths
of the Machulu study area
Earth Sci Inform
Fig. 5 Graph displaying absorption and reflectance features of
mineralized samples of Machulu area
(Fig. 5). The absorption features shown by group-2 samples
suggested that these were caused at 0.434 μm by jarosite, at
0.64 μm by epidote, at 1.4 μm by muscovite, at 1.937 μm by
albite, at 2.21 μm by kaolinite, and at 2.315 μm by amphibole
(Fig. 5). The group-3 samples showed spectral absorption fea-
tures at 0.434 μm, indicating jarosite, at 1.4 μm, indicating
paragonite, at 1.9 μm, indicating muscovite, at 2.2 μm, indi-
cating muscovite and lepidolite, and at 2.33 μm indicating
calcite (Fig. 5).
Astor area
Unaltered rocks Thirty-four (34) representative samples of
unaltered granitic, dioritic and gabbroic-dioritic rocks were
selected for the study. It was noted that each type of rock
showed specific spectral absorption and spectral reflectance
features at various wavelengths. The granitic rocks of the
study area showed spectral absorption features at wavelengths
of 1.413 and 1.92 μm, indicating muscovite, and at 2.205 μm,
indicating muscovite and lepidolite (Fig. 6). The dioritic rocks
showed absorption features at wavelengths of 1.412 μm, in-
dicating chlorite, and 2.205 μm, indicating calcite (Fig. 6). In
the gabbroic diorites, a very broad absorption spectrum at
wavelength 1.074 μm was noticed due to the presence of
olivine. No prominent absorption and reflectance spectra were
shown by gabbroic-dioritic rocks at other wavelengths,
Fig. 6 Graph showing depth of absorption and peaks of reflectance
features at different wavelengths of granitic, dioritic and gabbroic
dioritic (non-mineralized) rocks of study area Astor
Earth Sci Inform
presumably because the whole electromagnetic spectrum had
been absorbed by the minerals present in it (Fig. 6).
Sulfide-bearing altered rocks and quartz veins Twenty (20)
representative samples were collected from sulfide-bearing
quartz veins and associated altered rocks in order to study their
spectral absorption and spectral reflectance features at differ-
ent wavelengths. Rocks with similar spectra were organized in
groups, identified as group-A, group-B and group-C. For each
sample, analyses were performed at different angles in order
to identify spectral absorption and reflectance features at dif-
ferent sites. The absorption features of group-A were noted at
a wavelength of 0.836 μm, which is characteristic of the
copper-bearing phases such as malachite. The mineral
chrysocolla showed absorption features at a wavelength of
1.413 μm, while azurite showed absorption features at
1.92 μm. The minerals jarosite and limonite showed absorp-
tion features at wavelengths of 1.922 and 2.269 μm, respec-
tively (Fig. 7). The group-B samples showed different absorp-
tion minima and reflectance maxima at different wavelengths.
Absorption features noted at wavelengths of 0.5–0.56, 0.979–
0.983, 1.42, and 1.92 μm indicated the presence of hematite,
goethite, montmorillonite and jarosite, respectively (Fig. 7).
Group-C samples showed absorption features only at
0.413 μm, which is characteristic of quartz (Fig. 7).
Remote sensing
Landsat 8 data were used for mapping the unmapped study
area and for finding the sulfide-bearing alteration zones. The
Landsat 8 data were very helpful in mapping the geology in
both the study locations. The band ratio combination of 7/5–6/
3–4/3 proved very useful in differentiating various lithologies
in the Machulu area (Fig. 8a). As shown in Fig. 8a, the Hushe
gneisses were recognized by a reddish-brown color, diorites of
Ladakh batholith by a brownish and smoky color, Chalt and
Yasin group rocks by a smoky color, and metasediment by a
light blue color. Saltoro River, Hushe River, and Shyok River
were found to be very prominent in Fig. 8a. The Landsat 8
Fig. 7 Spectral absorption minima and spectral reflectance maxima at
various wavelengths of the mineralized rocks of Astor study area
data were also used for mapping different types of rock in the
Astor area, where the band ratio combination of 7/5–4/3–6/3
distinguished major geological units (Fig. 8b). As shown in
Fig. 8b, Chalt and Yasin group rocks were recognized by a
light green color, diorites of Kohistan Batholith by reddish-
brown, and the High Himalayan leucogranite and Sumayar
leucogranite by shiny blue.
Figure 9a shows ASTER band ratio (4/7–4/3–2/1) image
for Machulu area. This image clearly identifies diorites of
Ladakh batholith in bluish color. Chalt and Yasin group rock
can be seen in purple and greenish colors, rock alteration in
volcanics can be seen in brown color. Quaternary deposits are
seen in red color and Hushe gneisses in bluish color. Hushe
River and Shyok River can also be seen in the image.
Figure 9b shows ASTER band ratios for Astor area. In this
figure ASTER, band ratios 4/7, 4/3, 2/1 are displayed as Red-
Green-Blue, respectively. Figure 9b also show different litho-
logical units of the study area. The rocks of Chalt and Yasin
groups can be seen as alternate bands in blue and brown
colors, Kohistan batholith rocks are sky blue to dark blue
and the rocks of Nanga Parbat Syntaxis are light greenish
intermixing with red color (Fig. 9b). The glaciers give magen-
ta color.
Discussion
The spectral reflectances of minerals, rocks, and alteration
zones associated with particular mineralization are very help-
ful in mineral identification. These spectral reflectance studies
are also useful in selecting different bands for remote-sensing
analysis of unknown or inaccessible areas of high relief. Many
minerals show diagnostic (or diagnostically significant) ab-
sorption bands (Clark et al. 1990). The spectral features of
absorption and reflectance mainly depend on the chemical
composition, color, texture, and surface state of the rocks
and minerals (Bailin and Xingli 1991). The deeper the absorp-
tion band of the reflectance curve, the more abundant will be
the particular mineral in that rock (Clark 1999). In the wave-
length region of 0.4–2.5 μm, spectral reflectance is dominated
by hydroxyl ions (OHˉ), water (H2O) and ferrous ions (Fe+2)
that are present in the crystal structure of the rocks, and most
of the absorption bands are also caused by Fe+2, OHˉ, and
H2O (Vincent 1997). According to Vincent (1997), since the
absorption band of any mineral can very easily be influenced
by the interaction of an ion with the surrounding crystal struc-
ture, it varies from mineral to mineral. The position of the
absorption minima of an OHˉ ion depends on its position in
the crystal lattice, and most of the absorption features of the
OHˉ ions can be noted at 1.4, 2.2, and 2.3 μm (Hunt and
Salisbury 1970). At 1.4 and 1.9 μm, the absorption band nor-
mally indicates the presence of water, but if the absorption
 Earth Sci Inform

Fig. 8 a Landsat 8 image

showing band ratio 7/5–6/3–4/3
displayed as Red-Green-Blue,
respectively. Image is showing
different lithologies present in the
study area. b Landsat 8 image
showing band ratios 7/5, 4/3, 6/3
displayed as Red-Green-Blue,
respectively. The image
discriminate different lithologies
of this area. Brown color circles
show sample locations of the
study area

band is absent at 1.9 μm, then it shows that OHˉ ions are hydroxides. According to Hunt and Salisbury (1970), Fe+2
present in the mineral structure (Hunt and Salisbury 1970). produces a number of absorption bands that depend upon
Iron (Fe) is a ubiquitous element in almost all the earth’s the crystalline structure in which it is present, as well as on
material and usually occurs in the form of oxides and the impurities in that structure. The Fe+2 ion shows an
Earth Sci Inform

Fig. 9 a ASTER band ratio 4/7–

4/3–2/1 displayed as Red-Green-
Blue, respectively showing
lithological units and other
features for Machulu area. b
ASTER band ratios 4/7, 4/3, 2/1
displayed as Red-Green-Blue,
respectively. This image
discriminate different lithologies
of Astor area

absorption band near 1.1 μm, while various positive absorp- Kaolinite, which is the alteration product of aluminum silicate
tion minima are shown by ferric ion (Fe+3) at wavelength 0.4– minerals such as feldspar, shows depths of absorption features
0.9 μm (Bailin and Xingli 1991). Fayalite exhibits shallow around 1.4 μm due to the presence of OHˉ ions (Hunt and
reflectance at 1.0 μm caused by Fe +2 (Vincent 1997). Salisbury 1970). Clay minerals such as montmorillonite and

illite show strong water absorption at wavelengths of 1.4 μm
and 1.9 μm, while the depth of absorption at 2.2 μm is due to
OHˉ ions (Hunt and Salisbury 1970). However, according to
Clark et al. (1990), it is very difficult to distinguish between
the absorption bands of illite, muscovite and montmorillonite
in the 2.35–2.45 μm absorption bands. Due to the electronic
transition of Cu+2, a copper-bearing mineral such as malachite
shows a positive absorption edge between 0.4 and 0.6 μm and
a broad absorption band at a wavelength of 0.8 μm (Bailin and
Xingli 1991).
Considering the above mentioned causes, namely the var-
iations in the spectral features of absorption and reflectance,
the behavior of spectral features in the unaltered and altered
rocks of the study area has been synthesized here. It is noticed
that the unaltered rock samples of the study area show low
spectral reflectance as compared to those of altered rock sam-
ples. Some samples show high depth of absorption features at
particular wavelengths, indicating an abundance of particular
minerals in that sample. Jarosite, goethite, and hematite show
depth of absorption features in the range of 0.4–1.15 μm,
which can be due to the presence of Fe (Hunt and Salisbury
1970; Bailin and Xingli 1991). The minerals jarosite and li-
monite also show absorption minima at 2.2 and 2.26 μm,
which can be attributed to the presence of OH¯ ions. The
absorption features in the mineral antigorite at wavelengths
of 1.14, 1.935, and 1.974 μm are due either to the Fe+2 ion
or the OH¯ ion. The clay minerals illite, montmorillonite, and
muscovite show absorption features at the wavelengths of 1.4
and 1.9 μm due to the presence of OH¯ ions and H2O, while
muscovite also shows absorption minima at the wavelength of
2.2 μm due to the presence of the OH¯ ion (Hunt and
Salisbury 1970; Clark et al. 1990). Calcite shows a deep ab-
sorption feature at 2.32 μm. Plagioclase (anorthite) shows
absorption features at 1.4, 1.9, 2.24, and 2.33 μm, which can
be attributed to the OH¯ ion (Hunt and Salisbury 1970). A
slightly depressed absorption feature at wavelength 1.07 μm
in olivine can be caused by the presence of the Fe+2 ion
(Vincent 1997). A broad absorption feature in the range of
0.6–0.9 μm in malachite, a small absorption at 1.4 μm in
chrysocolla, and a deep absorption at 1.93 μm in azurite can
all be caused by the electronic transition of copper (Bailin and
Xingli 1991). Quartz also shows absorption minima at
0.4 μm.
Considering above mentioned absorption and reflectance
features of different minerals and rocks, certain bands were
selected for processing the Landsat images to differentiate
between various lithologies and, in turn, to locate sulfide-
bearing alteration zones, with the goal of finding the associat-
ed gold and base metals in the study area for exploration
purposes. Bands one through seven of Landsat 8 and bands
1–9 of ASTER data were selected for processing the remote-
sensing images. Various combinations of false colors and dif-
ferent ratios were applied to the Landsat images in order to
Earth Sci Inform
map the unmapped areas, and various alteration zones were
targeted for formulating exploration strategies for precious
and base metal mineralization. Very good results were obtain-
ed by applying band ratios in mapping the major geological
units in Astor and Machulu areas. For example, volcanic and
sediments of Chalt and Yasin group show iron leaching and
has clay minerals thus band ratio 4/3 in Landsat 8 and band
ratios of 2/1 and 4/3 in ASTER data helped discriminate these
rocks. Most of the granites in the study areas show absorption
features at 0.49 μm that is band 3 of Landsat 8 and band 2 of
the ASTER data. Spectral data for diorite samples show ab-
sorption features at various wavelengths, including 1.411 μm
and 1.914 μm, 2.247 μm, and 2.334 μm, indicating higher
amounts of feldspar thus band ratio 7/5 in Landsat 8 and 4/7 in
ASTER data were found very useful for mapping these rocks.
Although, band ratio images showed very prominent varia-
tions in their color, thus differentiating the various lithologies
of the study area. But, due to smaller size of alteration zones
satellite remote sensing turned out to be less useful.
Conclusions
Reflectance spectroscopic study of the unaltered and sulfide
bearing altered rocks of the two studies sites in northern Pa-
kistan helped identified several ore minerals like jarosite, goe-
thite, and hematite; also the copper-bearing phases malachite,
chrysocolla, and azurite were identified. Besides recognizing
mineralogical variations, reflectance spectroscopy helped in
selecting useful wavelengths for creating band ratios for
Landsat 8 and ASTER data for mapping major geological
units. Band ratios 7/5–4/3–6/3 for Landsat 8 and 4/7, 4/3,
2/1 for ASTER data proved best to differentiate between var-
ious lithologies in the study areas. Higher spatial and spectral
resolution data will be needed for mapping mineralized zones.
References
Ali L, Williamson BJ, Moon CJ, Shah MT, Khattak SA (2014)
Distribution of gold in different size fractions of stream sediments
as a guide to bedrock gold mineralization along the shyok suture
Zone and adjacent areas of the Kohistan Island Arc, Pakistan. Arab J
Geosci. doi:10.1007/s12517-014-1306-1
Austromineral (1976) Final report (feasibility study), Indus gold project:
submitted to Pakistan Mineral Development Corporation by
Austromineral, Vienna, Austria, 1–235
Austromineral (1978) Feasibility study (final report), Mineral exploration
and mining development, Chitral District: submitted to Sarhad
Development Authority, Peshawar Pakistan by Austromineral,
Vienna, Austria, 1–289
Bailin Y, Xingli Y (1991) Spectral reflectance features of rocks and ores
and their applications. Chin J Geochem 10(2):188–196
Earth Sci Inform
Beck RA, Burbank DW, Sercombe WJ, Riley GW, Barndt JK, Bery JR,
Afzal J, Khan AM, Jurgen H, Metje J, Cheema A, Shafique NA,
Lawrence RD, Khan MA (1995) Stratigraphic evidence for an early
collision between Northwest India and Asia. Nature 373:55–57
Biber K, Khan SD, Shah MT (2014) The source and fate of sediment and
mercury in Hunza River Basin, Northern Areas, Pakistan
Investigation of the source, fate, and transport of mercury in
Hunza River, Northern Areas, Pakistan. Hyrdol Process. doi:10.
1002/hyp.10175
Bignold SM, Treloar PJ (2003) Northward subduction of the Indian plate
beneath the Kohistan island arc, Pakistan Himalaya: new evidence
from isotopic data. J Geol Soc Lond 160:377–384
Chatterjee S, Goswami A, Scotese CR (2013) The longest voyage:
Tectonic, magmatic, and paleoclimatic evolution of the Indian plate
during its northward flight from Gondwana to Asia. Gondwana Res
23:238–268
Clark RN (1999) Chapter 1: spectroscopy of rocks and minerals, and
principles of spectroscopy, in manual of remote sensing. Remote
Sens Earth Sci 3:3–58
Clark RN, King VVT, Klejwa M, Swayze AG (1990) High spectral
resolution reflectance spectroscopy of minerals. J Geophys Res 95:
653–680
Clift PD, Hannigan R, Blustajn J, Draul AF (2002) Geochemical evolu-
tion of the Dras-Kohistan arc during collision with Eurasia: evidence
from the Ladakh Himalaya, India. Island Arc 11:255–273
Halfpenny R, Mazzucchelli RH (1999) Regional multi-element drainage
geochemistry in the Himalayan mountains, northern Pakistan. J
Geochem Explor 67:223–233
Hanson CR (1989) The northern suture in the Shigar Valley, Baltistan,
northern Pakistan. In: Malinconico LLJ, Lillie RJ (eds) Tectonics of
the Western Himalayas. Geological Society of America, Special
Paper 232:203–215
Hunt GR (1977) Spectral signatures of particulate minerals, in the visible
and near-infrared. Geophysics 42:501–513
Hunt GR, Salisbury JW (1970) Visible and near-infrared spectra of min-
erals and rocks: I. Silicate minerals. Mod Geol 1:283–300
Khan SD, Walker DJ, Hall S, Burke K, Shah MT, Stockli L (2009) Did
Kohistan-Ladakh island arc collide first with India? Geol Soc Am
Bull 121(3/4):366–384
Miandad S, Shah MT, Khan SD, Ahmad L (2014) Investigation for gold
and base metals mineralization and petrochemical characteristics of
the rocks of upper parts of Bagrot valley, Gilgit-Baltistan, Pakistan. J
Himal Earth Sci 47(2):29–48
Molnar P, Tapponnier P (1975) Cenozoic tectonics of Asia-Effects of a
continental collision. Science 86:1093–1102
Rehman AU, Shah MT, Khan SD, Ahmad L (2015) Investigation of gold
and base metals mineralization and the petrochemical studies of the
associated host rocks, Shagari Bala area, Skardu Gilgit-Baltistan,
Pakistan. J Himal Earth Sci 48(1):26–40
Rowley DB (1996) Age of initiation of collision between India and Asia:
a review of stratigraphic data. Earth Planet Sci Lett 145:1–3
Salisbury JW (1993) Mid-infrared spectroscopy laboratory data:
Chapter 4. In: Pieters CM, Englert PAJ (eds) Remote Geochemical
Analysis. Cambridge University Press, New York, pp 79–98
Shah MT, Khan H (2004) Exploration and extraction of placer gold in the
terraces of Bagrot valley, Gilgit, north Pakistan. Geol Bull Univ
Peshawar 37:27–40
Sheikh L, Shah MT, Khan SD, Ahmad L (2014) Investigation for gold
and base metals mineralization and petrochemical characteristics of
the rocks of Golo Das and surrounding areas, District Ghizar, Gilgit-
Baltistan, Pakistan. J Himal Earth Sci 47(2):15–27
Tahirkheli RAK (1974) Alluvial gold prospects in the North-west west
Pakistan. National Centre of Excellence in Geology, University of
Peshawar, Information Release No.7:1–50
Tahirkheli RAK (1979) Geology of Kohistan and adjoining Eurasian and
Indo-Pakistan continents, Pakistan. Geol Bull Univ Peshawar 11:1–
30
Treloar PJ, Petterson MG, Jan MQ, Sullivan MA (1996) A re-evaluation
of the stratigraphy and evolution of the Kohistan arc sequence,
Pakistan Himalaya: implications for magmatic and tectonic arc-
building processes. J Geol Soc Lond 153:681–693
Vincent RK (1997) Fundamentals of geological and environmental re-
mote sensing. Prentice-Hall, Inc., Upper Saddle River, pp 29–76
Yin A (2006) Cenozoic tectonic evolution of the Himalayan orogen as
constrained by along strike variation structural geometry, exhuma-
tion history, and foreland sedimentation. Earth Sci Rev 76:1–131
Yin A, Harrison TM (2000) Geologic evolution of the Himalayan-Tibetan
orogeny. Annu Rev Earth Planet Sci 28:211–280
Ziabrev SV, Aitchison JC, Abrajevitch A, Badengzhu DAM, Luo H
(2004) Bainang Terrane, Yarlung-Tsangpo suture, southern Tibet:
a record of intra-Tethyan subduction on the roof of the world. J
Geol Soc Lond 161(3):523–538