Hyperspectral Remote Sensing

CONTENTS
 Introduction
 Hyperspectral Data
 Hyperspectral Sensors
 Principle of imaging spectroscopy
 Multispectral v/s hyperspectral
 Processing of hyperspectral data
 Applications of Hyperspectral Remote sensing
 Applications in Geology

Dept. of Applied Geology Page 1

 Hyperspectral Remote Sensing

Introduction
Hyperspectral remote sensing is one of the advance technology which began in early 1980’s is one of the
most significant breakthroughs in remote sensing. It emerged as a promising technology in remote
sensing for studying earth surface materials by two ways spectrally and spatially. In this technology
imaging and spectroscopy is combined in a single system so this is also known as imaging spectroscopy.
This technology is developed by breaking a broad band from the visible and infra-red into hundreds of
spectral parts to obtain geochemical information from inaccessible planetary surfaces. Hyperspectral
remote sensing is able to provide a high level of performance in spectral and radiometric calibration
accuracy in the data sets. These high performing sensors data can be utilized for extracting information in
various quantitative and qualitative applications. The ample spectral information provided by
Hyperspectral data is able to identify and distinguish spectrally similar materials which enhance the
capability of distinguishing various ground objects in detail. Hyperspectral sensors collect information as
a series of narrow and contiguous wavelength bands at 10 to 20 nm intervals.

Hyperspectral Data
Hyperspectral remote sensing data sets are generally represented as a data cube with spatial information
collected in XY plane, and spectral information represented in the Z direction. The hyperspectral data sets
are composed of about 100 to200 narrow and contiguous wavelength band at bandwidths of about 5-10
nm. The spectra for a single pixel in hyperspectral data appears similar like a laboratory quality spectra
which can be used for understanding the spectral characteristics of the material. Due to the ample spectral
information provided by hyperspectral data it is easy to identify and distinguish spectrally similar
materials.

Figure: Hyperspectral Remote Sensing

Hyperspectral Sensors
In the hyperspectral field there are two types of systems that take images: on aircraft and on satellites.
Most hyperspectral sensors are mounted on aerial platforms and less on the satellite. In general, the
spectral range within which hyperspectral sensors on aircraft work is 380-12700 nm and for those on the
satellites is 400-14400 nm. Most hyperspectral sensors record the reflected radiation in a series of bands
with narrow and continuous wavelength. The number and width of bands varies from one system to
another, i.e. in the range of 1-288, with widths ranging from 2-2000 nm. Unlike the sensors on aircraft,
sensors on satellites have the capacity to provide global coverage at regular intervals. Further is presented
Dept. of Applied Geology Page 2
 Hyperspectral Remote Sensing

a comparison between the AVIRIS air sensor and the Hyperion satellite sensor considered to be
representative for the two types of systems. The Hyperion EO-1 sensor was launched in November 2000
by NASA with the purpose of taking hyperspectral images from space in order to create mineralogical
mapping. Hyperion is a hyperspectral satellite sensor which works in the spectral range 0.4-2.50 µm with
242 bands which have a spectral resolution of about 10 nm and a spatial resolution of 30 meters, the data
is taken from an altitude of 705 km.

Sensor Spectral Coverage No. of bands Spatial Country

(nm) Resolution(m)
GERIS(Geophysical 400-1000 24 1-10 USA
Environment 1400-1800 7
Research Imaging 2000-2500 32
Spectrometer II)
AVIRIS (Airborne 380-2500 220 5-20 USA
Visible Infrared
Imaging
Spectrometer)
CASI (Compact 400-800 288 30 Canada
Airborne Imaging
Spectrometer)
DAIS (Digital 400-1200 72 1-10 Europe
Airborne Imaging 1500-1800
Spectrometer) 2000-2500
HYDICE 400-2500 10.2 3 USA
( Hyperspectral
Data Image
Collection
Experiment)
HyMAP 400-2500 16 3-5 Australia
AsiaEAGLE 400-900 5 <1

Spaceborne Hyperspectral sensors

Sensor Spectral No. of bands Spatial Resolution Agency
Coverage(nm)
Moderate Resolution 400-800 32 250-1000 NASA
Imaging
Spectrometer(MODIS)-
AQUA

MODIS- TERA 800-1455 36 250-1000

MERIS (Medium 410-1050 15 Ocean:1040x1200 ESA
Resolution Imaging Land & Coast:
Spectrometer) 260x300
Hyperion on EO-1 400-2500 220 3 NASA
CHRIS(Compact High 438-1035 18-64 18-36 ESA
Resolution Imaging
Spectrometer on

Dept. of Applied Geology Page 3

 Hyperspectral Remote Sensing

PROBA-1)
HySI(Hyperspectral 400-950 64 550 ISRO
Imager) on IMS-1

Extraterrestrial Hyperspectral sensors

Sensor Spectral No. ob bands Spatial Resolution Agency
Coverage(nm)
Chandrayaan-1 400-920 64 80 ISRO
HySI
Chandrayaan-1 400-3000 86 70-140 ISRO
M3 (Moon
Mineralogy
Mapper)
OMEGA 360-5100 300-4000 NASA
CRISM (Compact 362-3920 545 15.7-19.7 NASA
Reconnaissance
Imaging
Spectrometers for
Mars)

Hyperion is a push-broom instrument that takes pictures with a radiometric resolution of 8 bits, the band
having a width of 7.5 km and being perpendicular on the movement of the satellite. The system used for
taking images is formed of two spectrometers: one working in the visible/near infrared (VNIR) (0.4…1.0
µm) and one in shortwave infrared (SWIR) (0.9…2.5 µm). The data are calibrated using both the
radiation measured before the mission and when the images are taken. Airborne Visible Sensor/Infrared
Imaging Spectrometer (AVIRIS) are new in terms of hyperspectral systems attached to planes. The
AVIRIS sensor, developed by NASA/Jet Propulsion Laboratory (JPL), is working in bands of 224, with a
spectral resolution of about 10 nm and covering the spectral range from 0.40 to 2.50 µm. The sensor is a
Whiskbroom system that uses a scanning system for acquiring data on the transverse direction of
advancement. Four off-axis double-pass Schmidt spectrometers capture light from fore optics using
optical fiber and send it to four linear panels, one for each spectrometer, which have a strong sensitivity in
the range 0.4…0.7 µm, 0.7… 1.2 µm, 1.2…1.8 µm and 1.8…2.5 µm. AVIRIS sensor takes images from
an altitude of 20 km with a spatial resolution of 20 meters, from a band whose width is of 10.5 kilometers.
Starting with 1998, the sensor is mounted on a Twin Otter aircraft flying at low altitude, taking pictures
with a spatial resolution ranging between 2 and 4 meters.

Principle of Imaging Spectroscopy

As Hyperspectral Remote sensing technology is also known as Imaging Spectroscopy which is considered
to be as combination of three following photonic technologies:

(i) Conventional imaging

(ii) Spectroscopy
(iii) Radiometry

Above three technologies are used to produce images for which a spectral signature is associated with
each pixel. The datasets produced by Hyperspectral imager is in the form of three dimensional data cube
in which two dimensions represent spatial information and the third dimension represents spectral

Dept. of Applied Geology Page 4

 Hyperspectral Remote Sensing

information. The values recorded by Spectral Imager Instrument can be converted via proper calibration,
to radiometric quantities that are related to the scene phenomenology (e.g. radiance, reflectance,
emissivity, etc). This technology is having capability to support various applications by providing a link
to spatial and spectral analytical models, spectral libraries, etc.

Figure1. Relationship among Radiometric, Spectrometric and imaging Technique

Spectroscopy depends on the pretext that the different materials are different because of the difference in
their constituents and structure and because of that they interact differently with light so they appear
different. The aim of Imaging Spectroscopy is to understand the Earth’s surface through the detailed
analysis of its reflected light, exploiting subtle variations in surface composition and structure in support
of real world requirements. For spectroscopic study, Hyperspectral data sets provide ample spectral detail
to discern the subtle differences in color distributions from Earth surface materials. Because Earth’s
surface is populated with the molecules of solids and liquids and having characteristics spectral features
generally wider than some tens of nanometers, which establishes a practical definition for the maximum
spectral band size for a Hyperspectral data set.

The reflectance spectra of most of the Earth’s surface materials contain characteristic or diagnostic
absorption features in the spectral range of 400 to 2500nm. Since these diagnostic features are typically of
very narrow spectral appearance, those surface materials can be identified directly, if the spectrum is
sampled at sufficiently high spectral resolution which becomes possible using imaging spectrometers.
There are three types of main absorption features found generally in the spectral range of 400 to02500 nm
regions which should be understood to realize the requirement of Hyperspectral imaging system.

a) Charge transfer absorption: these types of absorptions are caused by light at certain
wavelengths causing electrons to be transferred between atoms and generally occur in the visible
region of the spectrum, and for example: Fe+3 and Fe+2. Light at the proper wavelength causes an
electron to be transferred from a Fe+2 atoms to a Fe+3 atoms and due to that rusty objects appear
red. Detection of this type of absorption is easy as they are quiet broad, so it is possible to detect
those using conventional multispectral sensors. As there is overlap among the absorptions caused
by different atoms, so Hyperspectral sensors are required to tell them apart.
b) Electron transmission absorption: in atoms with an appropriate electron shell, light at the
proper wavelength can bump electrons into different positions in the shell. These absorptions tend

Dept. of Applied Geology Page 5

 Hyperspectral Remote Sensing

to be narrower than the charge transfer absorptions and the type of atom and the position and
variety of its neighbors controls the wavelength of the absorptions. This feature is especially
useful in geology, where the arrangements of atoms in a mineral are very well defined. Since
subtle variations in the position of the band centre is important, it is necessary to have many
narrowly spaced bands to take advantage of this feature.
c) Variational absorptions: when the light of the same wavelength as the molecule (or a part of a
molecule) strikes the molecule, it causes the molecule (or part of the molecule) to vibrate. This
leads to light absorption. In general these absorptions are very much narrow, although their
widths and depths vary. Many of the absorptions seen in the 0.4 to 0.25 µm region actually
originate at longer wavelength. Most of these absorptions can be detected with a multi spectral
sensor.

Actual detection of materials is dependent on the spectral coverage, spectral resolution and signal to noise
ratio of the spectrometer, the abundance of the material and the strength of absorption features for that
material in the wavelength region measured. Most natural Earth surface materials have diagnostic
absorption features in 400 to 2500 nm range of the reflected spectrum. Since the diagnostic features for
each material is apparent over very narrow spectral bands, differences between materials can only be
identified if the spectrum is sampled at a sufficiently high resolution. The benefit of Hyperspectral remote
sensing is that the information about the ground objects can be recorded in a very narrow spectral range,
hence minute alterations can be mapped.

Multispectral v/s Hyperspectral

Multispectral data sets are produced by sensors which record reflected electromagnetic energy within
some specific sections or broad bands of the electromagnetic spectrum. These sensors usually produce 3
to10 numbers of spectral bands which ranges from visible to near infrared region. However, the spectral
resolution and mineral discrimination power is low. Examples of multispectral satellite sensor are
Landsat, Spot and IRS satellites.

Hyperspectral sensors measure energy in narrower and more numerous bands than multispectral sensors.
Hyperspectral data contains 100s or more narrow contiguous spectral bands. The numerous narrow bands
of hyperspectral sensors provide contiguous spectral measurement across the entire electromagnetic
spectrum and therefore are more sensitive to subtle variations in reflected energy.

Images produced from hyperspectral sensors contain much more data than images from multispectral
sensors and have a greater potential to detect differences among land and water features. Hyperspectral
sensors are having capability to detect and distinguish individual absorption bands in mineral deposits,
vegetation and man-made materials. This discrimination is achieved by spectral sampling at
approximately 10nm intervals across the spectrum.

Multispectral images can be used to map forested areas, whereas hyperspectral images can be used to map
tree species within the forest.

Dept. of Applied Geology Page 6

 Hyperspectral Remote Sensing

Figure: Comparison between hyperspectral and multispectral sensors

Processing of Hyperspectral Data

Recent advances in sensor technology have led to the development of Hyperspectral sensors capable of
collecting imagery containing several hundred bands over the spectrum. However, the increase in the
number of bands is both a blessing and a curse. The large number of bands provides the opportunity for
more materials to be discriminated by their respective spectral response. However, this large number of
bands is the characteristic which leads to complexity in analysis techniques.

One difficulty in working with Hyperspectral data is to understand the difference associated with working
in n-dimensional space.

Data Pre-Processing Techniques

The processing of hyperspectral data sets are more complex as it contains large number of spectral bands
which results in larger data volume. The hyperspectral data quality is also affected by the random sensor
errors and noise which causes low signal to noise ratio. These data sets are also affected by atmospheric
errors, so pre-processing of hyperspectral data is required; pre-processing includes sensor and
atmospheric error removal and image rectification.

Sensor error correction: most of the hyperspectral sensors like Hyperion are pushbroom sensors in
which poorly calibrated detectors produce vertical lines on the image. Due to poor calibration the bad
lines are having different values than the neighboring pixels, either the values are constant or lower than
the neighboring values. These bad lines can be corrected by replacing their DN values with the average
DN values of their immediate left and right neighboring pixels because of the high spatial correlation.

Dept. of Applied Geology Page 7

 Hyperspectral Remote Sensing

Atmospheric correction: the atmosphere scatters some of the electromagnetic energy which travels from
the sun to the earth’s surface and from the earth’s surface to the sensor. Therefore, the electromagnetic
energy received at the sensor may be more or less than that due to reflectance from the earth’s surface
alone. Atmospheric correction attempts to minimize these effects on image spectra. Atmospheric
correction is traditionally considered to be indispensible before quantitative image analysis using
hyperspectral data. Various atmospheric correction algorithms have been developed to calculate
concentrations of atmospheric gases directly from hyperspectral data. Atmospheric correction is divided
into two types: Relative and Absolute methods.

1. Relative Method is divided into three types

a) Flat Field correction

b) Empirical line correction

c) Internal apparent relative reflectance correction

2. Absolute atmospheric correction: this method is based on some atmospheric correction models
which require the information regarding the atmospheric condition, altitude, geometry between
sun and the satellite, aerosol level, water absorption, time of acquisition of the image and more
details. The absolute atmospheric correction methods have the advantage over other methods that
these can be run under any atmospheric condition. Some of them are

a) FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes); an

atmospheric correction modeling tool in ENVI for retrieving spectral reflectance from
hyperspectral radiance images. FLASH incorporates the MODTRAN 4 radiation transfer
model to compensate for atmospheric effects.

b) ATCOR (Atmospheric and topographic CORrection): ATCOR algorithm has been

developed in the last decade in two different types (ATCOR 2 & ATCOR 3) which are
created by Dr. Richter of the German Aerospace Center-DLR. For nearly horizontal surface
or flat terrain, ATCOR 2 is designed for rugged topographical surface; hence a Digital
Elevation Model (DEM) is used in the ATCOR 3 algorithm for atmospheric correction.

c) ATREM (Atmospheric REMoval Program): ATREM is software developed by the

University of Colorado for retrieving scaled surface reflectance from hyperspectral data using
a radiative transfer model.

d) ACORN (Atmospheric CORrection Now): the atmospheric correction now (ACRON)

software package provides an atmospheric correction of Hyperspectral and multispectral data
measured in the spectral range from 350 to 2500nm. It is also based on MODTRAN 4
radiative transfer code.

Data Dimensionality Reduction

Although, the hyperspectral data are voluminous and multidimensional, now-a-days with the availability
of advanced computing systems that possess high speed processors and enormous storage power, data
volume is no longer a constraint. The problem lies in the data redundancy that needs to be removed to
obtain the bands with maximum information.

Dept. of Applied Geology Page 8

 Hyperspectral Remote Sensing

Much of the data does not add to the inherent information content for a particular application, even
though it often helps in discovering that information; it contains redundancies. The data recorded by
hyperspectral sensors often have substantial overlap of information content over the bands of data
recorded for a given pixel. In such cases, not all of the data are needed to characterize a pixel properly,
although redundant data may be different for different applications. Data redundancy can take two forms;
spatial and spectral. Since hyperspectral imagery has more spectral concern, one way of viewing spectral
redundancy in hyperspectral data is to form the correlation matrix for an image; the correlation matrix can
be derived from the covariance matrix. High correlation between band pairs indicates high degree of
redundancy. For example, PCT (Principle Component Transformation) assist in removing redundancy
since decorrelation followed by a discarding of low variance components amounts to redundancy
reduction.

Principle Component Analysis (PCA)

PCA transform multidimensional image data into a new, uncorrelated co-ordinate system or vector space.
It produces a space, in which the data have maximum variance along its first axis, the next largest
variance along a second mutually orthogonal axis and so on. Sometimes even the lower order PC’s may
contain valuable information. The later principle components would be expected, in general to show little
variance. These could be considered therefore to contribute little to separability and could be ignored,
thereby reducing the essential dimensionality of the classification speed. Stated differently, the purpose of
this process is to compress all the information contained in an original n-band data set into fewer than n
“new bands” or components. The components are then used in line of the original data. These
transformations may be applied as a preprocessing procedure prior to automated classification process of
the data.

Radiometric Correction
Hyperspectral imaging sensors collect radiance data from either airborne or space borne platforms which
must be converted to apparent surface reflectance before analysis techniques can take place. Atmospheric
correction techniques have been developed that use the data themselves to remove spectral atmospheric
transmission and scattered path radiance. There are seven gases in the Earth’s atmosphere that produce
observable absorption features in the 0.4-2.5 micron range. They are water vapor, carbon dioxide, ozone,
nitrous oxide, carbon monoxide methane and oxygen.

Approximately half of the 0.4 - 2.5 micron spectrum is affected by gaseous absorption is illustrated below
in Figure 1. For this reason, the ATREM 3.0 (Atmosphere Removal Program) developed by the Center
from the Study of Earth from Space (CSES) at the University of Colorado can be used to remove the
effects of the atmosphere from AVIRIS or HYDICE imagery. ATREM is available via anonymous ftp at
cses.colorado.edu from the pub/atrem directory.

Dept. of Applied Geology Page 9

 Hyperspectral Remote Sensing

The ATREM software was developed to determine the scaled surface reflectance from hyperspectral
imagery from both AVIRIS and HYDICE sensors. The atmospheric scattering used by ATREM is
modeled after the MODTRAN 5S code. The ATREM software assumes that the surface is horizontal and
has a Lambertian reflectance. If topography is known, then the scaled surface reflectance can be
converted into real surface reflectance.

The ATREM model is a good approximation to radiometric correction of the imagery. However,
calibration of the ATREM surface reflectance to in situ measurements should improve the final results. A
by-product from the ATREM software is an image of the columnar water vapor which was removed from
the input hyperspectral data. The two figures below represent an AVIRIS frame prior to the ATREM
correction and a water vapor scene removed from an AVIRIS scene which was acquired over the
Kennedy Space Center on March 23, 1996. The images show a significant amount of water vapor
removed from the imagery which causes attenuation of the upwelling radiance.

Dept. of Applied Geology Page 10

 Hyperspectral Remote Sensing

Columnar water vapor image removed from AVIRIS

Original AVIRIS data over KSC (Bands 20, 29, 40)
data using ATREM program.

Minimum Noise Fraction (MNF) Transformation

While hyperspectral imagery is capable of providing a continuous spectrum ranging from 0.4 to 2.5
microns (in the case of AVIRIS) for a given pixel, it also generates a vast amount of data required for
processing and analysis. Due to the nature of hyperspectral imagery (i.e. narrow wavebands), much of the
data in the 0.4-2.5 micron spectrum is redundant.

A minimum noise fraction (MNF) transformation is used to reduce the dimensionality of the
hyperspectral data by segregating the noise in the data. The MNF transform is a linear transformation
which is essentially two cascaded Principal Components Analysis (PCA) transformations. The first
transformation decorrelates and rescales the noise in the data. This results in transformed data in which
the noise has unit variance and no band to band correlations. [ENVI] The second transformation is a
standard PCA of the noise-whitened data.

For this particular example, an AVIRIS frame over the Kennedy Space Center was radiometrically
corrected using ATREM and a MNF transformation was performed on the ATREM-corrected imagery. In
this particular frame, the first 14 eigenvectors of the MNF transformation contain coherent information
which can be used for further processing.

Dept. of Applied Geology Page 11

 Hyperspectral Remote Sensing

Eigenvectors 1, 2, & 3 of MNF Transform Data Eigenvectors 6, 9, & 12 of MNF Transform Data

Pixel Purity Index

The Pixel Purity Index (PPI) is a processing technique designed to determine which pixels are the most
spectrally unique or pure. Due to the large amount of data, PPI is usually performed on MNF data which
has been reduced to coherent images. The most spectrally pure pixels occur when there is mixing of
endmembers. The PPI is computed by continually projecting n-dimensional scatter plots onto a random
vector. The extreme pixels for each projection are recorded and the total number of hits is stored into an
image. These pixels are excellent candidates for selecting end-members which can be used in subsequent
processing.

End Member Extraction

Theoretically the existing pure features in mixed pixels are referred to as endmembers. Selection and
identification of spectral endmembers in an image is the key point to success of the linear spectral mixing
model. A set of endmembers should allow the description of all spectral variability for all pixels. Two
different approaches have generally been used to define endmembers in a mixing model:

 Use of the existing library of reflectance spectra

 Extraction of the purest pixels from the image data itself

Endmembers resulting through the first option are denoted as known endmembers whereas the second
option results in derived endmembers. Because of the difficulties of access to spectral library or field
measurement of spectral properties of land cover types of interest, endmembers data of known ground
cover types can be extracted from the hyperspectral data.

Dept. of Applied Geology Page 12

 Hyperspectral Remote Sensing

Applications of Hyperspectral Remote Sensing

Hyperspectral data has been used to identify and distinguish spectrally similar materials having
characteristic reflectance spectra. Due to the capability of distinguishing various ground objects in detail,
hyperspectral data sets are able to detect and map a wide variety of materials.

 Mineral Targeting: Spectral reflectance in visible and near-infrared offers a rapid and
inexpensive technique for determining the mineralogy of samples and obtaining information on
chemical composition.
 Soils: obtaining quantitative information about soil chemistry, its genetic and fertility
classification. Study of soil parameters such as organic matter, soil moisture, particle size
distribution, iron oxide content, soil structure, etc. The hyperspectral data with improved
radiometric and spatial resolution will help in deriving an improved vegetation/soil indices that
will maximize sensitivity to plant biophysical parameters, increase sensitivity to plant biophysical
parameters, increase sensitivity to the vegetation signal and normalize atmosphere and ground
contamination noise influence.
 Vegetation: study of species diversity, environmental stress, physiological features such as
photosynthetic activity, plant productivity, canopy biochemistry, biomass and plant transpiration.
Also for evaluation of vegetation stress, nutrient stress, moisture stress and crop growth models.
 Atmosphere: study of atmospheric parameters such as clouds, aerosol conditions and water
vapour monitoring, large scale atmospheric variations as a result of environmental change.
 Oceanography: measurement of photosynthetic potential by detection of phytoplankton,
detection of yellow substance and detection of suspended matter. It also helps in investigations of
water quality, monitoring coastal erosion.
 Snow and Ice: spatial distribution of various types of snow cover, surface albedo and snow water
equivalent. Calculation of energy balance of a snow pack, estimation of snow properties- snow
grain size, snow depth and liquid water content.
 Oil Spills: when oil spills in an area affected by wind, waves and tides, a rapid and assessment of
the damage can help to maximize the cleanup efforts. Environmentally sensitive areas can be
targeted for the protection and cleanup and the long term damage can be minimized. The
sequence images of the oil can guide efforts in real time by providing relative concentrations and
accurate locations.

Applications in Geology
In this section, we discuss a few applications of hyperspectral sensing data in the solar reflection region
(VIS-NIR-SWIR). Some airbome hyperspectral sensors also carry spectral bands in the thermal-IR region
which are useful for the identification of rock-forming silicate minerals. As these bands are similar to
those of TIMS, the method of data processing and interpretation would be similar to that of
TIMS/ASTER.

A number of applications of hyperspectra1 sensing data in mineralogical studies have been reported.
There are two main approaches: (a) mineralogical identification and (b) mineral species quantification.

1. Skarns: Skarns or tactites are calcium-bearing calc–silicate rocks. Skarns are most often
formed at the contact zone between intrusions of granitic magma bodies and carbonate
sedimentary rocks such as limestone and dolostone. Hot fluids derived from the granitic
magma are rich in silica, iron, aluminium, and magnesium. These fluids mix in the contact
zone, dissolve calcium-rich carbonate rocks, and convert the host carbonate rock to skarn
deposits in a metamorphic process called metasomatism. Skarns are typically coarse-grained

Dept. of Applied Geology Page 13

 Hyperspectral Remote Sensing

and zoned bodies. They constitute the principal source of tungsten, and also contain minor
deposits of tin, cobalt, gold, arsenic, silver and many other metals. Besides, several industrial
minerals are also associated with skarns, chiefly asbestos, graphite, wollastonite and
magnesite.

limestone has an absorption band at 2.35 µm and dolomite at 2.32 µm . Using this spectral
characteristics and Geoscan MK 11 imagery, Windeier and Lyon (1991) delineated a
dolomitized limestone unit in the Ludwig skarn, Nevada.

2. Clay mineralogy: The Cuprite mining district, Nevada, is one of the most thoroughly studied
geological test sites. Van der Meer and Bakker (1998) and van der Meer (1999) analysed A
VIRIS data of the Cuprite area in order to identify distribution of various clay minerals:
kaolinite, alunite and buddingtonite. Figure 11.25 shows a comparison of the spectra of the
three clay minerals obtained from AVIRIS and laboratory samples. Based on the spectral data
comparison (crosscorrelogram approach), van der Meer and Bakker (1998) mapped the
distribution of these minerals in the Cuprite mining district.

3. Kaolinite content image: In central parts of Germany, there occur huge overburden dumps
due to lignite open-cast mining, which was carried out on a large scale during the last
century. Information on composition of dumps and its spatial variation is important in order

Dept. of Applied Geology Page 14

 Hyperspectral Remote Sensing

to understand the hydro-geochemical effect of the dumps. Kaolinite forms an important

alteration mineral in the overburden dumps.

4. Hydrocarbon Exploration: Hyperspectral sensing can form a very important tool for
detecting onshore hydrocarbon micro-seepages . It is not only fast and low-cost, but can also
detect even those traps which may be overlooked by seismic methods. Micro-seepages of
hydrocarbons can result in several changes on the surface, such as:

 Reduction of ferric oxide

 Conversion of clay and feldspars

 Increase of carbonate content

 Anomalous blue shift of the red edge of vegetation spectra

The above transformations on the surface are amenable to detection on hyperspectral remote
sensing data. Van der Meer et al. (2001) give an example of the detection of bleached beds
due to hydrocarbon seepage using Probel/HyMAP data

Conclusion:

Hyperspectral remote sensing is an advanced technique in remote sensing which depicts the earth surface
material with number of discrete and narrow spectral bands. This topic provided knowledge on different
hyper spectral sensors, data acquisition and processing methods. In addition to this, the information on
application of hyper spectral data in various fields for characterizing and mapping the surface materials is
provided with a special focus on the geological applications.

References:

Remote Sensing Geology – Ravi P. Gupta

Remote Sensing of the Environment-John R. Jenson

https://nrsc.gov.in/sites/all/pdf/SPIE%20APRS%20Tutorial_Hyperspectral%20RS_Vinay%20Kumar.pdf

http://www.csr.utexas.edu/projects/rs/hrs/process.html

Dept. of Applied Geology Page 15