ERS186: Environmental Remote Sensing

1
ERS186Environmental Remote Sensing

• Lecture 6
• Geology

2
Overview

• Applications
• Geology (Rocks and Minerals)
• Physical Principles
• Scattering
• Absorption
• Photon energy
• Spectrum (reflectance and emission)
• Sensors
• AVIRIS
• ASTER

3
The Question

• Most of geological remote sensing asks the
  following question
• For a reflectance curve (spectrum) either
  determined in the lab, with a field spectrometer,
  or with a remote sensor, what is the chemical
  composition and structure of these chemicals
  within the field of view of the instrument?
• Or in other words What kind of rock am I
  looking at?

4
Classification

• One possible end product of remote sensing
  imagery is a discrete class for each pixel in an
  image.
• A similar goal is to determine the composition
  of classes within each pixel (since most pixels
  are mixtures of materials) and/or the structural
  elements within a pixel (e.g. grain sizes).

5
Basic Definitions

• Rock an assemblage of minerals held together by
  some sort of cement (silica or calcium carbonate)
• Hapkes equation to model the reflectance (r?)
  from an exposed rock
• µ0cosine of angle of incident light, µcosine of
  angle of emitted light, gphase angle, waverage
  scattering albedo from rock or mineral,
  Bgbackscatter function, Pgaverage single
  particle phase function, Hfunction for isotropic
  scatterers.
• Whew! The point of all of this is that with
  knowledge of known optical constants of various
  minerals and the angle of incident and emitted
  light, we can MODEL the reflectance of a rock
  with mixed grain sizes and minerals!

6
Light and Rocks

• What does light do when it hits a rock?
• Scattered
• Reflects off of the grains either away from the
  surface or onto other grains
• Refracted through a grain onto other grains
• Absorbed by a grain
• Where does the light come from?
• Sun (reflected)
• The mineral itself (emitted)

7
Reflection and Absorption

• Complex index of refraction (m)
• m n-jK n is real part of index, j (-1)1/2
  and K extinction coefficient
• Beers Law
• I I0e-kx, I observed light intensity, I0
  original light intensity, k absorption
  coefficient (k 4K/?) and x is the distance
  travelled through the medium
• Consequences
• As distance increases through a medium, more
  photons will be intercepted.
• The rate of interception is higher with shorter
  wavelengths
• Fresnel equation
• R (n-1)2K2/(n1)2K2, Rreflection of
  light normally incident onto a plane surface
• Reflectance factor
• ?? L?/Lcalib, a spectral signature (aka
  spectrum, aka spectral reflectance curve) is the
  plot of ?? vs. ?

8
Identifying Minerals

• Different types of minerals absorb and scatter
  incident energy differently for different
  wavelengths of light!!!
• These differences in absorption and scattering
  for different wavelengths can be used to identify
  the minerals.
• We examine the maxima and minima of spectral
  reflectance curves minima are caused by
  molecular absorption, and we call these
  absorption features or absorption bands.

9
What causes absorption features?

• Electronic processes
• Crystal field effect an electron is moved from a
  lower level to a higher level by the absorption
  of a photon with the exact energy difference
  between the two states (remember, Qh?). Occurs
  in Ni, Cr, Co, Fe, etc. and absorption bands are
  typically small.
• Charge transfer absorptions caused when an
  electron is transferred to another ion or ligand
  due to the absorption of a photon. They cause
  large absorptions in the UV extending into the
  visible. This is the cause of red color of iron
  oxide.
• Conduction bands photon of a specific energy
  causes a shift of an electron into the electronic
  lattice of certain materials (dielectrics, not
  metals). Occurs in the visible to NIR regions.
  This is the cause of the yellow color of sulfur.
• Color centers irradiation of an imperfect
  crystal (one with defects) causes an electron to
  shift into the defect.

10
Electronic Processes
Crystal field effect absorption caused by Fe2.
Fe 29 has 53.65 FeO, Fe 91 has 7.93 FeO.
Charge transfer absorption caused by Fe2. Fe2O3
(hematite) and FeOOH (geothite).
Conduction bands caused by S and HgS.
11
What causes absorption features?

• Vibrational processes
• Bonds in a molecule vibrate, the frequency is
  dependent on the type of bond and the atom
  masses.
• Certain materials have important vibrational
  absorption bands water, hydroxyl, carbonates,
  phosphates, borates, arsenates, vanadates.

12
Identifying Minerals

• We can use all of these absorption features to
  determine the chemical composition of a spectral
  reflectance curve.

13
Mixtures

• 4 types of mixtures
• Linear (areal) mixture materials in the field of
  view of the detector are optically separated, so
  the reflectance at each wavelength is the
  fraction of each material times its particular
  reflectance for that wavelength
• Intimate mixture different materials are in
  close contact (e.g. mineral grains in soil or
  rock), the combined reflectance is non-linear
  combination of the individual reflectances.
• Coatings materials coating other materials.
  Each layer will have a different optical
  thickness with differing optical properties.
• Molecular mixtures mixing occurring on a
  molecular level (e.g. solutions). Resultant
  reflectance is non-linear.

14
Mixtures
15
Grain Size

• Scattering occurs on the surface of grains,
  whereas absorption occurs within grains (in
  accordance to Beers Law).
• Surface to volume ratio comes into play smaller
  grains have higher surface to volume ratio, so
  usually more scattering occurs in smaller grains
  vs. larger grains, and reflectance is typically
  higher in the VNIR.

16
Continuum and Band Depth

• Absorption features defined by continuum and the
  depth of the absorption
• Think of the continuum as the continuation of the
  reflectance curve if there was no absorption
• The depth of the absorption feature is defined
  as
• D1-Rb/Rc, Rb is the reflectance at the bottom
  of the absorption feature and Rc is the
  reflectance of the continuum at the same
  wavelength.
• Key point the depth of an absorption feature is
  related to the abundance of the absorber and the
  grain size of the material.

17
Spectral Libraries

• Collections of high radiometric and spectral
  resolution spectrum of various materials.
• Collected spectrum can be compared to these
  spectrum to identify them.
• Important to note data from spectrometers is
  collected in radiance, but must be converted to
  reflectance factor to compare to other samples.
• With an adequate spectral library, geological
  remote sensing can be done without field data
  (consider planetary geology).

18
Spectral Resolution

• Absorption features are typically very narrow (lt
  20 nm), so narrow band widths are necessary.
• Many important band widths are also fairly
  shallow, so high radiometric resolution and SNR
  is also necessary.

19
AVIRIS
20
Cuprite, NV

• True color (LANDSAT TM bands).
• We are interested in mapping the minerals in the
  non-vegetated regions.

21
Cuprite, NV

• Derived from the electronic absorption features
  (0.4 to 1.2 microns)
• Fe2 and Fe3 bearing minerals
• Grain size can be determined using saturated
  bands
• The absorption features are broad, so specific
  mineralogy is more difficult to determine

22
Cuprite, NV

• Derived from the vibrational absorption features
  (2 to 2.5 microns)
• OH, CO3 and SO4 bearing minerals
• Can use absorption depths to determine the amount
  of mineral in a pixel

23
Mapping Mine Waste
California Gulch Superfund Site, Leadville, CO
Pyrite (and Fe-bearing secondary minerals) are
indicative of acidic mine waste (Swayze et al.
1999).
24
September 11, 2001

• Almost immediately following 9/11, AVIRIS was
  flown over Manhattan.
• AVIRIS was used to determine the amount of
  asbestos (chrysolite and amphibole) released when
  the towers fell.

25
September 11, 2001
26
ASTER

• With well placed bands at known absorption
  regions, and multiple TIR bands, ASTER can
  provide relatively high quality information about
  mineral compositions.
• 6 SWIR bands band 6 centered at the clay
  absorption feature and band 8 at the carbonate
  absorption feature
• 5 TIR bands bands 10, 11, and 12 at sulfate and
  silica absorption features

27
ASTER
Cuprite, NV classification of ASTER data (SWIR
bands 4,6,8 on left). Bluekaolinite,
redalunite, light greencalcite, dark
greenalunitekaolinite, cyanmontmorillonite,
purpleunaltered, yellowsilica or dickite.
28
Emissivity

• Another way we can identify minerals is through
  their emissivity spectrum (as opposed to their
  reflectance spectrum), located primarily in the
  TIR range.
• Emissivity (e) is the ratio between the true
  radiant flux from an object (Mr) and the
  hypothesized blackbody emission at the same
  temperature (Mb).
• e Mr/Mb
• Mb sT4 (Stefan-Boltzmann law)

29
Thermal Emission Spectroscopy