Lab Exercise #4: Spectral Signature Concepts
Purpose:
To provide exposure to: (1) acquisition of spectral radiometric data, (2)
calibration to radiance values, (3) spectral signature interpretation, and (4)
visualization of results in graphic and tabular formats. This will enable you
to gain an awareness of several fundamental aspects of reflected radiation.
Materials:
• Data values previously collected with an Exotech 4 band radiometer with filters
for Landsat Thematic Mapper bands 1 (blue), 2 (green), 3 (red) and 4 (NIR).
• Spreadsheet program to perform calculations and construct graphs, or:
• A basic calculator, data summary tables and graphing paper
Email your completed word document and excel spreadsheet to: kwarkentin@sdsu.edu
with your names and lab 4 in the subject line.
A user may be able to choose the proper bands and filters for a given remote sensing task
by collecting and analyzing reflectance data for various targets, which have been or are to
be imaged in a given scene, and then by identifying spectral bands which maximize class
separability. Theoretically, the greater contrast in reflectance between two imaged
objects, the easier it should be to distinguish between them. The easier objects are to
distinguish, the greater is the potential for fast, accurate interpretation.
Table 1. Landsat Thematic Mapper & Exotech Radiometer Bands
Band
Spectral Bands (m) Exotech bands in bold
1
0.45 - 0.52
blue
2
0.52 -0.60
green
3
0.63 -0 .69
red
4
0.76 -0 .90
near IR
5
1.55 - 1.75
mid IR
7
2.08 - 2.35
mid IR
6
10.4 - 12.5
thermal
A four channel Exotech radiometer was used several years ago to collect raw spectral
radiometric measurements for general surface material types around campus. The
radiometer measures the radiation reflected from the surface as a voltage. You will be
provided with radiant exitance values using the calibration data in Table 2. The voltages
recorded on the data sheets for each band of each target were multiplied by the calibration
for the band, field-of-view (FOV) and sensor gains used in the conversion program.
Radiant exitance (Wm-2) = Voltage * Calibration factor * 10exponent
Calculate radiant exitance and enter below for each target. A 15 degree FOV was used
for all measurements. Average the beginning and ending panel radiant exitance.
You may choose to do all table-based calculations in the Excel spreadsheet which is
provided as part of this exercise, following the general workflow described here.
Table 2: Radiant Exitance (Wm-2)
Target
average panel
asphalt
concrete
grass
shrub
soil
shaded grass
shaded concrete
Band 1
140
22
93
6
1
Band 2
152
35
86
11
3
Band 3
95
12
43
5
1
Band 4
154
45
35
103
72
3
2
49
7
5
21
65
2
13
39
26
8
The halon coated panel used to measure the incident radiation at the surface is not a perfect,
i.e., neither 100% nor an isotropic (Lambertian) reflector. Therefore, an adjustment must be
applied to the panel measurements. The average panel radiant exitance that you calculated
above must be divided by a factor taken from Table 3 below for the time that most closely
matches when your observations were taken (around 2:30 PM). The magnitude of the
adjustment is dependent on the solar zenith angle which is a function of the time of day, day
of the year and latitude. For your calculations, use the 2:30 PM adjustment factors.
Table 3 - Halon Panel Adjustment Factors
Band 1
Band 2
Band 3
.935458
.939679
.934222
.932881
.937191
.931803
Band 4
.919215
.916819
2:30
2:45
3:00
3:15
3:30
.92994
.926623
.922896
.918706
.913986
.934368
.931202
.927663
.923701
.919248
.929077
.926039
.922662
.918895
.91467
.914118
.91111
.907772
.90406
.899914
3:45
4:00
.908651
.902605
.914222
.90853
.909903
.904497
.895261
.890013
Time (PDT)
2:00
2:15
2
Enter the adjusted panel values below (you can use the excel spreadsheet for the
calculations) :
Avg. Adjusted Panel Values = Average panel radiant exitance panel adjustment
(Divide the average panel radiant exitance by the correct panel adjustment factors)
Band 1
Band 2
Band 3
Band 4
Average Adjusted
Panel Values
Now, calculate the spectral reflectance factor for each band of the seven targets (numbers
and calculations can be stored and submitted in the spreadsheet). The spectral
reflectance factor is the target radiant exitance divided by the adjusted average panel
radiant exitance. Express the reflectance as a decimal fraction out to four significant
digits beyond the decimal point.
Reflectance = Target exitance (Wm-2) Adj. avg. panel exitance (Wm-2)
Reflectance
Material
Band 1
Band 2
Asphalt
Concrete
Grass
Shrub
Soil
Shaded Grass
Shaded Concrete
3
Band 3
Band 4
ANALYSIS
1. Plot reflectance for each target as a point with the bands on the X axis. Connect the
points for each band. Create one graph for each target, and a final graph depicting all of
the targets on one graph (you should create 8 graphs). You may find the graph more
interpretable if you don’t use point markers, such that each target is depicted as a line. For
the final graph, construct a colored bar graph, with bands 1-4 on the X-axis, and values
for each target on the Y-axis, color-labelled by material type. Use colors that are
reasonable for each target. Include axis titles on each graph.
Save the graphs in the excel spreadsheet. You do not need to copy them to the word doc.
2. Using the matrix below, determine the bands of electromagnetic spectrum which allow
the optimum discrimination of any one target from another (fill in the best 2 bands). (Use
sunlit targets, i.e., not shaded, only for this analysis.) Only fill in the white boxes.
Optimum band selection for discriminating between target pairwise combinations.
Asphalt
Concrete
Grass
Shrub
Soil
Asphalt
Concrete
Grass
Shrub
Soil
3. Select the bands which are optimum for discriminating among the three general
classes of vegetation, asphalt, and concrete/soil (fill in the best band or combination of
bands in the blanks in the matrix). Only fill in the white boxes. Discuss the rationale for
your decisions.
Optimum band(s) selection for discriminating between three general classes of targets
Vegetation
Asphalt
Vegetation
Asphalt
Concrete/Soil
4
Concrete/Soil
Explanation:
4. Discuss the reflectance calculated for the two shaded targets. How do the shaded
reflectance compare to one another and to the sunlit targets of the same material types?
5. Why is it important to investigate the nature of wavelength reflectance from targets
when planning remote sensing missions?
6. Describe the utility of "multi-spectral" approaches for separating land cover types, i.e.,
potential advantages of using more than two spectral bands?
5
Table 1. Landsat Thematic Mapper Bands
Spectral Bands (mm)
Band
1
0.45 - 0.52
2
0.52 -0.60
3
0.63 -0 .69
4
0.76 -0 .90
5
1.55 - 1.75
7
2.08 - 2.35
6
10.4 - 12.5
Exotech bands in bold
blue
green
red
near IR
mid IR
mid IR
thermal
Average Adjusted Panel Values:
1
2
3
4
Table 2: Radiant Exitance (Wm-2)
Target
average panel
asphalt
concrete
grass
shrub
soil
shaded grass
shaded concrete
Band 1
140
22
93
6
1
3
2
49
Table 3 - Halon Panel Adjustment Factors
Band 2
152
35
86
11
3
7
5
21
Band 1
Band 2
Band 3
150.547347 162.6765 102.2517
Band 3
95
12
43
5
1
65
2
13
Band 4
168.468
Band 4
154
45
35
103
72
39
26
8
Time (PDT)
Band 1
2:00
2:15
2:30
2:45
3:00
3:15
3:30
3:45
4:00
0.93546
0.93288
0.92994
0.92662
0.9229
0.91871
0.91399
0.90865
0.90261
Material
Band 1
0.146133
0.617746
0.039855
0.006642
0.019927
0.013285
0.325479
Asphalt
Concrete
Grass
Shrub
Soil
Shaded Grass
Shaded Concrete
Example Data:
l Adjustment Factors
Band 2
Band 3
Band 4
Bands:
Values 1:
Values 2:
0.93968 0.93422 0.91922
0.93719 0.9318 0.91682
0.93437 0.92908 0.91412
0.9312 0.92604 0.91111
0.92766 0.92266 0.90777
0.9237
0.9189 0.90406
0.91925 0.91467 0.89991
0.91422 0.9099 0.89526
0.90853 0.9045 0.89001
Band 4
0.21515
0.52866
0.06762
0.01844
0.04303
0.03074
0.12909
0.11736
0.42053
0.0489
0.00978
0.63569
0.01956
0.12714
0.26711
0.20775
0.61139
0.42738
0.2315
0.15433
0.04749
0.7
Reflectence Values
0.6
0.5
0.4
0.3
0.2
0.1
0
1
2
3
4
3
4
Bands
Concrete
0.7
0.6
Reflectence Values
Band 3
2
2
2
Grass
0.5
0.4
0.3
0.2
0.1
0
1
2
Bands
Soil
0.3
Reflectence Values
Band 2
1
0.5
1
0.25
0.2
0.15
0.1
0.05
0.05
0
1
2
3
Bands
4
3
4
0.5
4
2.5
5
Asphalt
0.35
0.3
0.3
Reflectence Values
Reflectence Values
0.25
0.2
0.15
0.1
0.25
0.2
0.15
0.1
0.05
0.05
0
0
1
4
2
3
4
Bands
Shrub
0.7
0.45
0.6
0.35
0.3
0.5
0.25
0.2
0.15
0.1
0.05
0
1
2
3
Bands
4
Shaded Grasst
4
Reflectence Values
Reflectence Values
0.4
0.4
0.3
0.2
0.1
0.18
0.16
0.14
Axie
0.12
0.1
0.08
0.06
0.04
0.02
0
0.02
4
0
1
2
3
Bands
4
Shaded Concrete
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
1
2
3
4
Bands
Final Chart
Band 1
Band 2
Band 3
Band 4
Bands
Asphalt
Concrete
Grass
Shrub
Soil
Shaded Grass
Shaded Concrete
Band 4
haded Concrete
EMR – Target/Surface Interactions
Partitioning of Energy at Surface
◼
Radiant flux at the surface is partitioned
among:
◼
Absorption
◼
Transmission
◼
Reflection
Radiation Budget Equation (cont.)
◼
Dimensionless ratios:
Absorptance:
al = Fabsorbed / Fil
Transmittance: tl = Ftransmitted / Fil
Reflectance:
◼
rl = Freflected / Fil
al + tl + rl = Fil = 1
Radiation Budget Equation (cont.)
◼
Proportion of energy absorbed/transmitted/reflected
will vary from target-to-target
◼
◼
◼
◼
Material type
Material condition
For a given target, proportion absorbed, transmitted,
and reflected energy will vary with wavelength
Ability to distinguish between targets or measure
phenomena
Hemispherical al , tl , rl
◼
Radiometric quantities based on the amount of radiant
energy incident to a surface from ANY angle in a
hemisphere; direct and diffuse sources
Percent Reflectance
◼
Percent reflectance =
rrl = (Freflected / Fil ) * 100
◼
Widely used in RS research to describe surface/target
reflectance characteristics
Properties Affecting Target Reflectivity
◼
Target/surface
◼
◼
◼
◼
EMR
◼
◼
◼
◼
Moisture/electrical - nature of penetration & absorption
Micro-roughness - concrete, leaves, soil texture
Macro-roughness - terrain → shadowing, re-direction of light
Wavelength - for same target, amount of reflectance varies with l
Polarization – orientation of incident light may affect amount of
reflectance (microwave)
Angle of Incidence – angle of incoming light
Interaction of properties determines target radiance
detected by the sensor
Types of Reflection
◼
Two types of reflection
◼
◼
◼
Specular
Diffuse (Lambertian)
Type determined as a function of:
◼
◼
Snell’s Law
Rayleigh criterion of surface roughness
Types of Reflection (cont.)
Uniform
reflection
Smooth
surface
Rough
surface
Snell’s Law (refraction)
Angle of incidence
Angle of reflection
Zenith
Air
Leaf
Smooth
surface
Angle of transmission
Specular Reflection
◼
Rayleigh criterion: surface is SMOOTH in comparison
to the incident wavelength if:
h ≤ l / (8 cos qi)
where:
◼
h = height variations above surface
l = wavelength
qi = angle of incidence
◼
Blue light (0.4 mm) : h ≤ 0.07 mm
◼
Microwave (1000 mm) : h ≤ 176 mm
Angle of incidence = angle of reflection
Diffuse (Lambertian) Reflection
◼
Rayleigh criterion: surface is ROUGH in comparison to
the incident wavelength if:
h > l / (8 cos qi)
◼
◼
◼
Blue light (0.4 mm) : h > 0.07 mm
Microwave (1000 mm) : h > 176 mm
Random scattering in all directions
Reflection (cont.)
◼
◼
◼
Geometry of reflection = ƒ (surface roughness)
Surface roughness = ƒ (wavelength of incident
energy)
Rocky terrain
◼
◼
Smooth to longer microwave wavelengths
Fine sand
◼
Rough to shorter visible wavelengths
Bidirectional Reflectance Distribution Function
(BRDF)
◼
Energy detected by sensor is a function of:
◼
◼
Solar geometry AND
Sensor viewing geometry
Angle of Incidence/Observer Angles
Spectral Signature Concept
◼
◼
Describes spectral reflectance of a target at different
wavelengths of EMR
Spectral reflectance curve - graphs reflectance
response as a function of wavelength
◼
Key to separating and identifying objects
◼
Selection of optimum wavelength bands
Typical Spectral Reflectance Curves
More Spectral Reflectance Curves
A
Spectral resolution
B
Factors Affecting Leaf Reflectance
Hemispherical Reflectance, transmittance, and Absorption Characteristics of
Big Bluestem Grass
Jensen, 2000
Vegetation Reflectance – by wavelength
Absorption Spectra of
Chlorophyll a and b,
b-carotene, Pycoerythrin,
and Phycocyanin Pigments
Chlorophyll a peak absorption is at
0.43 and 0.66 mm.
Chlorophyll b peak absorption is at
0.45 and 0.65 mm.
Optimum chlorophyll absorption
windows are:
0.45 - 0.52 mm and 0.63 - 0.69 mm
Spectral Reflectance
Characteristics of
Sweetgum Leaves
(Liquidambar
styraciflua L.)
Jensen, 2000
Spectral Reflectance
Characteristics of
Selected Areas of
Blackjack Oak
Leaves
Jensen, 2000
Cross-section Through
A Hypothetical and
Real Leaf Revealing
the Major Structural
Components that
Determine the Spectral
Reflectance
of Vegetation
Hypothetical
Example of
Additive
Reflectance
from A Canopy
with Two Leaf
Layers
Jensen, 2000
Effect of Leaf Water Content on
Reflectance
Dominant Factors Controlling Leaf
Reflectance
Water
absorption
bands:
0.97 mm
1.19 mm
1.45 mm
1.94 mm
2.70 mm
Jensen, 2000
Effect of Stress on Vegetation Reflectance
Spectra for Urban Materials
Measurement of Spectral Signatures
◼
Radiometer
◼
◼
◼
◼
Records radiance from an object in specific wavelengths
May record data for fixed bands or wavelength region
Spectro-radiometer measures near-continuous spectra
Reflectance Calculations
◼
◼
Incident radiance is determined by taking measurements
over barium sulfate or halon panel (reference panel)
Bidirectional reflectance factor is calculated by dividing
spectral radiance from object by panel measurement for
given wavelength
Handheld
Data Logger
Avoid self-shadowing
or reflectance
Non-imaging
Goniometer in Operation at North Inlet, SC
Boom Mounted
Imaging Radiometer
◼
Imaging Radiometer: radiometer + “scanning”
capability – builds a 2D image
Airborne Visible
Infrared Imaging
Spectrometer
(AVIRIS) Datacube of
Sullivan’s Island
Obtained on October
26, 1998
Imaging Spectrometer Data of Healthy Green Vegetation in the
San Luis Valley of Colorado Obtained on September 3, 1993
Using AVIRIS
224 channels each 10 nm wide with 20 x 20 m pixels
Jensen, 2000
Purchase answer to see full
attachment