Airborne Remote Sensing Patterns And Processes Of Land And Water
This document was uploaded by user and they confirmed that they have the permission to share it. If you are author or own the copyright of this book, please report to us by using this DMCA report form. Report DMCA
Overview
Download & View Airborne Remote Sensing Patterns And Processes Of Land And Water as PDF for free.
More details
- Words: 1,545
- Pages: 43
2009 Student Airborne Research Mission
Airborne Remote Sensing Patterns and Processes of Land and Water
Professor Susan Ustin: [email protected] Graduate Student Assistant: Shawn Kefauver [email protected] NASA-UND National Suborbital Education and Research Center
Remote Sensing Principles
Electromagnetic Spectrum
Visible Thermal IR Reflected Solar IR
Remote Sensing Principles
Solar Spectral Radiance
Top of Atmosphere Solar Irradiance Surface Spectral Radiance
How does light interact with matter to create the measurement?
Plant Functions
Pigment absorption is the dominant process in visible; Scattering is the dominant process in near-infrared; Water absorption is increasingly important with wavelength in the mid-infrared.
Plant Functions
MASTER Instrument Characteristics
MASTER Spectral Band Positions (Vis-MWIR)
0.01 to 6 μm
Band Centers vs. Atmospheric Features
Radiation Budget Equation Φ iλ
= Φ reflected + Φ transmitted + Φ absorbed
Where Φ iλ = total flux incident on surface, in watts or J/s Hemispherical reflectance
rλ = Φ reflected / Φ iλ
Hemispherical transmittance
τ λ= Φ transmitted / Φ iλ
Hemispherical absorptance
α λ = Φ absorbed / Φ iλ
Fate of Photons Interacting with a Surface
rλ + τ λ +α λ = 1
Summary of Light Interactions with Matter
Refraction
rλ τλ ελ α
= reflection = transmission = emission = absorption
r λ + τ λ + ελ + α λ = 1
10
The Law of Specular Reflectance Scattering:
Incidence angle = Exitance angle
Specular Scattering
Forward scatter direction (specular direction)
Specular reflection from water
Leaf biochemistry: A typical leaf cell contains: • Photosynthetic pigments (chloroplasts) • chlorophyll a and b •carotenoids: β-carotene, xanthophylls • Other pigments (cytoplasm) • anthocyanins, flavons • brown pigments • etc.
• water (vacuole): 90-95% • dry matter (cell walls): 5-10% • cellulose: 15-30% • hemicellulose: 10-30% • proteins: 10-20% • lignin: 5-15% • starch: 0.2-2.7% • sugar • etc.
Absorption by foliar pigments
Chlorophyll α and b b
a
β-carotene
chlorophyll a
chlorophyll b
Anthocyanin
β-carotene
Chemical Structures of Pigments are Known
anthocyanin + glucose
http://harvardforest.fas.harvard.edu/research/leaves/
Plant Functions
■
453
■
■
430 662 410
652
Senescence sequence of a leaf
16
Optical Properties
Reflectance Changes as Water Content Declines
Water Volume: Equivalent water thickness
g Wf − Wd EWT 2 = A cm
Dry matter content
g Wd DM = 2 A cm
Wf = Fresh weight Wd = Dry weight A = Leaf Area
Soil Texture 100 (% ) Cla y
60
40
sandy clay
30
Clay
90
80
sandy loam
60 70 Sand (%)
read 40
silty clay
50 60
silty clay loam
clay loam
70
%)
sandy clay loam
loamy Sand sand
100
30
50
read
20
• Different fractions are identified as different soil-texture classes.
20
80 70
10
10
( Silt
• Proportion of sand, silt and clay in a soil (or horizon); usually calculated as % of soil by weight for each type of particle
90
Loam
80
silt loam
90 Silt
50
40 read
30
20
10
100
100 50 90 80 Silt Sand
70
% Reflectance
Percent Reflectance
Basic Dry Soil Spectra
30 60 50
Silt
Sand
40
1030 20 10 0
0.5 0.7
0.9
1.1
1.3 1.5 1.7 1.9 Wavelength (µm)
2.1
2.3
2.5
Reflectance ~increases with wavelength from the visible, to the mid- infrared portion of the spectrum
Soil Moisture and Texture 60 50
Sand Sand Sand
0 – 4% moisture content
40 30
5 – 12%
20
22 – 32%
10
a.
0 0.5 60 50
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2.3
2.5
Clay Clay
Clay
2 – 6%
40 30 20
35 – 40%
10
b.
0 0.5
0.7
0.9
1.1
1.3 1.5 1.7 1.9 Wavelength (µm)
2.1
2.3
2.5
Clays hold water more ‘tightly’ than sand. Thus, water absorption bands in a clay spectrum are more prominent than in a sand spectrum. Hyperspectral data can be used to quantify these absorption features.
Soil Organic Matter
Organic matter is a strong absorber of EMR in the VIS, so more organic matter produces darker soils (lower reflectance). Note change in shape from concave to convex.
Soil Reflectance and Moisture Content
Wilting Point 1.5MPa
Field Capacity 0.03MPa Bowers and Hanks (1965)
Soil Influences in Refletcance Data over San Joaquin Valley, CA Irrigation Texture
7-1
8-2 8-3
Location of N Treatment Richard Plant, UC Davis
Kirchhoff’s Law: A theorem (based on lab observations) states that at a given temperature, energy is absorbed and radiated at the maximum possible rate per unit area, for each wavelength.
Fails at high frequencies (very short wavelengths)
San Francisco Bay Estuary and Pacific Ocean Coastline
Color Infrared
Sediment Load
Surface Water Temperature
Light Penetration of Water by Wavelength and Depth
Blue
Green
Red
NIR
Light Penetration of Water
Cozumel Island
Palancar Reef SPOT Band 1 (0.5 0.59 µm) green
Caribbean Sea SPOT Band 2 (0.61 0.68 µm) red
SPOT Band 3 (0.79 0.89 µm) NIR Jensen, 2000
Upwelling along the northern and central California coast, SeaWIFS 6 Oct. 2002
chlorophyll-bearing phytoplankton
28
% reflectance of clear and algae-laden water
% reflectance of algaeladen water at suspended sediment concentrations ranging from 0 500 mg/l
Han, 1997; Jensen, 2000
5 4.5
1,000 mg/l
4
clay
250 200 150
3.5 Percent Reflectance
Clear Water with Different Levels of Suspended Clay and Silt Soil Sediments
Clayey soil
3
300
100
2.5
50
2 clear water
1.5 1 0.5
a.
0 400
450
500
550
600 650 700 Wavelength (nm)
14
silt
12
550 500 450
Silty soil
Percent Reflectance
10
300 250 200
8
150
850
900
400
Reflectance peak shifts toward longer wavelengths as more suspended sediment is added
600
100
6
50
4 clear water
2 0 400
800
1,000 mg/l
350
b.
750
450
500
550
600 650 700 Wavelength (nm)
750
800
850
900
Lodhi et al., 1997; Jensen, 2000
Mapping Aquatic Plant Species True Color Image
The Sacramento Delta Detection based on spectral differences between water and plants
Map of Submersed Aquatic Weeds
Color Infrared Image
Map of Aquatic Plant Species
Emergent Aquatic Species
Rhode Island
Pennywort
Pennywort
Water Hyacinth
Water Hyacinth
MASTER Spectral Band Positions (Thermal IR)
Band Centers vs. Atmospheric Features
Blackbody Radiation Curves for Several Objects Emitting Energy at different Temperatures, including the Sun and Earth
Wein’s Displacement Law λ max = k/ T k= 2898 μm K T = K (deg. Kelvin)
Sun = ~ 6000 K Earth = ~300 K
The mean Earth Temp. = 300K
λmax
k = T
2898 µm K = = 9.67 µm ≈ 300 K
Thermal Emissions In the Solar Region
True Color (visible)
500nm Band
1000 nm Band
2000 nm Band
Willow Fire, CA Sept. 01, 1999
λmax
k = T
2898 µm K = =9 .67 µm 2.07 300 K 1400
≈
Radiance,Radiance(uW/cm^2/nm/sr) W cm-2 nm Sr. -1
6.00
5.00
AVIRIS Estimate Residual
4.00
Test fit for spectrum G Temperature Estimate=984K 1.48% of the area
3.00
2.00
1.00
0.00 400
0.60
700
1000
0.800 1.00
1300
1600
1900
2200
2500
1.20 1.40 1.60 1.80 2.00
Wavelength (nm)
How Much Energy is Emitted by an Object? Stefan-Boltzmann Law (for ideal blackbodies) Mb is simplification and derived from Planck’s equation as Mb, λ = σT4 Total radiated energy (j or watts/m2) by a blackbody σ = 5.6697 x 10-8 W m-2 K-4 μm K What is a “perfect blackbody” ? How about near-perfect blackbody? What about objects that don’t emit at all wavelengths?
Emissivity Emissivity (ε): the ratio between the true radiant flux from an object (Mr) and the blackbody emission at the same temperature (Mb). ελ = Mr/Mb = (Trad/Tkin)4 Mλ= εσT4 Incorporating Emissivity (ε) allows calculation of Radiant flux of nonblackbody materials
Death Valley, CA Landsat natural color (left) and TIMS (center
The range in Digital numbers (DNs) is the Radiometric Resolution
What Is A Digital Image? Pixel Digital Number (DN)
70
53
41
64
84
85
81
88
91
87
79
77
45
38
59
77
84
86
85
85
80
82
69
44
32
45
72
86
82
78
88
79
86
87
65
40
41
75
79
78
93
86
93
106
106
84
56
43
58
75
104
104
100
101
95
91
83
51
39
56
105
110
97
88
84
85
87
77
59
44
96
103
89
79
79
75
77
79
74
72
87
93
97
90
82
76
70
67
61
71
79 81
DN are typically ranges from 0 to 255; 0 to 511; 0 to 1023, etc. These ranges are binary scales: 28=256; 29=512; 210=1024.
Current instruments typically have 214 What What your computer you see…sees… or 216 DN = white By convention, 0=black 255 81
88
97
93
85
78
74
70
72
75
78
85
94
97
92
84
80
72
Spectral Bands in Image Dataset Reflected Light from 3 bands
Light Source Each pixel is measured by an individual detector.
B3 B2 B1
Grayscale vs. RGB Grayscale represents the data in the band as an image where the variation in values is represented by the different intensity tones. … RGB (“Red”, “Green”, “Blue”) images are composites of 3 bands, corresponding to the red, green and blue phosphors on a monitor. Computer monitor colors are additive, meaning maximum DN of red + green + blue = white.
Airborne Remote Sensing Patterns and Processes of Land and Water
Professor Susan Ustin: [email protected] Graduate Student Assistant: Shawn Kefauver [email protected] NASA-UND National Suborbital Education and Research Center
Remote Sensing Principles
Electromagnetic Spectrum
Visible Thermal IR Reflected Solar IR
Remote Sensing Principles
Solar Spectral Radiance
Top of Atmosphere Solar Irradiance Surface Spectral Radiance
How does light interact with matter to create the measurement?
Plant Functions
Pigment absorption is the dominant process in visible; Scattering is the dominant process in near-infrared; Water absorption is increasingly important with wavelength in the mid-infrared.
Plant Functions
MASTER Instrument Characteristics
MASTER Spectral Band Positions (Vis-MWIR)
0.01 to 6 μm
Band Centers vs. Atmospheric Features
Radiation Budget Equation Φ iλ
= Φ reflected + Φ transmitted + Φ absorbed
Where Φ iλ = total flux incident on surface, in watts or J/s Hemispherical reflectance
rλ = Φ reflected / Φ iλ
Hemispherical transmittance
τ λ= Φ transmitted / Φ iλ
Hemispherical absorptance
α λ = Φ absorbed / Φ iλ
Fate of Photons Interacting with a Surface
rλ + τ λ +α λ = 1
Summary of Light Interactions with Matter
Refraction
rλ τλ ελ α
= reflection = transmission = emission = absorption
r λ + τ λ + ελ + α λ = 1
10
The Law of Specular Reflectance Scattering:
Incidence angle = Exitance angle
Specular Scattering
Forward scatter direction (specular direction)
Specular reflection from water
Leaf biochemistry: A typical leaf cell contains: • Photosynthetic pigments (chloroplasts) • chlorophyll a and b •carotenoids: β-carotene, xanthophylls • Other pigments (cytoplasm) • anthocyanins, flavons • brown pigments • etc.
• water (vacuole): 90-95% • dry matter (cell walls): 5-10% • cellulose: 15-30% • hemicellulose: 10-30% • proteins: 10-20% • lignin: 5-15% • starch: 0.2-2.7% • sugar • etc.
Absorption by foliar pigments
Chlorophyll α and b b
a
β-carotene
chlorophyll a
chlorophyll b
Anthocyanin
β-carotene
Chemical Structures of Pigments are Known
anthocyanin + glucose
http://harvardforest.fas.harvard.edu/research/leaves/
Plant Functions
■
453
■
■
430 662 410
652
Senescence sequence of a leaf
16
Optical Properties
Reflectance Changes as Water Content Declines
Water Volume: Equivalent water thickness
g Wf − Wd EWT 2 = A cm
Dry matter content
g Wd DM = 2 A cm
Wf = Fresh weight Wd = Dry weight A = Leaf Area
Soil Texture 100 (% ) Cla y
60
40
sandy clay
30
Clay
90
80
sandy loam
60 70 Sand (%)
read 40
silty clay
50 60
silty clay loam
clay loam
70
%)
sandy clay loam
loamy Sand sand
100
30
50
read
20
• Different fractions are identified as different soil-texture classes.
20
80 70
10
10
( Silt
• Proportion of sand, silt and clay in a soil (or horizon); usually calculated as % of soil by weight for each type of particle
90
Loam
80
silt loam
90 Silt
50
40 read
30
20
10
100
100 50 90 80 Silt Sand
70
% Reflectance
Percent Reflectance
Basic Dry Soil Spectra
30 60 50
Silt
Sand
40
1030 20 10 0
0.5 0.7
0.9
1.1
1.3 1.5 1.7 1.9 Wavelength (µm)
2.1
2.3
2.5
Reflectance ~increases with wavelength from the visible, to the mid- infrared portion of the spectrum
Soil Moisture and Texture 60 50
Sand Sand Sand
0 – 4% moisture content
40 30
5 – 12%
20
22 – 32%
10
a.
0 0.5 60 50
0.7
0.9
1.1
1.3
1.5
1.7
1.9
2.1
2.3
2.5
Clay Clay
Clay
2 – 6%
40 30 20
35 – 40%
10
b.
0 0.5
0.7
0.9
1.1
1.3 1.5 1.7 1.9 Wavelength (µm)
2.1
2.3
2.5
Clays hold water more ‘tightly’ than sand. Thus, water absorption bands in a clay spectrum are more prominent than in a sand spectrum. Hyperspectral data can be used to quantify these absorption features.
Soil Organic Matter
Organic matter is a strong absorber of EMR in the VIS, so more organic matter produces darker soils (lower reflectance). Note change in shape from concave to convex.
Soil Reflectance and Moisture Content
Wilting Point 1.5MPa
Field Capacity 0.03MPa Bowers and Hanks (1965)
Soil Influences in Refletcance Data over San Joaquin Valley, CA Irrigation Texture
7-1
8-2 8-3
Location of N Treatment Richard Plant, UC Davis
Kirchhoff’s Law: A theorem (based on lab observations) states that at a given temperature, energy is absorbed and radiated at the maximum possible rate per unit area, for each wavelength.
Fails at high frequencies (very short wavelengths)
San Francisco Bay Estuary and Pacific Ocean Coastline
Color Infrared
Sediment Load
Surface Water Temperature
Light Penetration of Water by Wavelength and Depth
Blue
Green
Red
NIR
Light Penetration of Water
Cozumel Island
Palancar Reef SPOT Band 1 (0.5 0.59 µm) green
Caribbean Sea SPOT Band 2 (0.61 0.68 µm) red
SPOT Band 3 (0.79 0.89 µm) NIR Jensen, 2000
Upwelling along the northern and central California coast, SeaWIFS 6 Oct. 2002
chlorophyll-bearing phytoplankton
28
% reflectance of clear and algae-laden water
% reflectance of algaeladen water at suspended sediment concentrations ranging from 0 500 mg/l
Han, 1997; Jensen, 2000
5 4.5
1,000 mg/l
4
clay
250 200 150
3.5 Percent Reflectance
Clear Water with Different Levels of Suspended Clay and Silt Soil Sediments
Clayey soil
3
300
100
2.5
50
2 clear water
1.5 1 0.5
a.
0 400
450
500
550
600 650 700 Wavelength (nm)
14
silt
12
550 500 450
Silty soil
Percent Reflectance
10
300 250 200
8
150
850
900
400
Reflectance peak shifts toward longer wavelengths as more suspended sediment is added
600
100
6
50
4 clear water
2 0 400
800
1,000 mg/l
350
b.
750
450
500
550
600 650 700 Wavelength (nm)
750
800
850
900
Lodhi et al., 1997; Jensen, 2000
Mapping Aquatic Plant Species True Color Image
The Sacramento Delta Detection based on spectral differences between water and plants
Map of Submersed Aquatic Weeds
Color Infrared Image
Map of Aquatic Plant Species
Emergent Aquatic Species
Rhode Island
Pennywort
Pennywort
Water Hyacinth
Water Hyacinth
MASTER Spectral Band Positions (Thermal IR)
Band Centers vs. Atmospheric Features
Blackbody Radiation Curves for Several Objects Emitting Energy at different Temperatures, including the Sun and Earth
Wein’s Displacement Law λ max = k/ T k= 2898 μm K T = K (deg. Kelvin)
Sun = ~ 6000 K Earth = ~300 K
The mean Earth Temp. = 300K
λmax
k = T
2898 µm K = = 9.67 µm ≈ 300 K
Thermal Emissions In the Solar Region
True Color (visible)
500nm Band
1000 nm Band
2000 nm Band
Willow Fire, CA Sept. 01, 1999
λmax
k = T
2898 µm K = =9 .67 µm 2.07 300 K 1400
≈
Radiance,Radiance(uW/cm^2/nm/sr) W cm-2 nm Sr. -1
6.00
5.00
AVIRIS Estimate Residual
4.00
Test fit for spectrum G Temperature Estimate=984K 1.48% of the area
3.00
2.00
1.00
0.00 400
0.60
700
1000
0.800 1.00
1300
1600
1900
2200
2500
1.20 1.40 1.60 1.80 2.00
Wavelength (nm)
How Much Energy is Emitted by an Object? Stefan-Boltzmann Law (for ideal blackbodies) Mb is simplification and derived from Planck’s equation as Mb, λ = σT4 Total radiated energy (j or watts/m2) by a blackbody σ = 5.6697 x 10-8 W m-2 K-4 μm K What is a “perfect blackbody” ? How about near-perfect blackbody? What about objects that don’t emit at all wavelengths?
Emissivity Emissivity (ε): the ratio between the true radiant flux from an object (Mr) and the blackbody emission at the same temperature (Mb). ελ = Mr/Mb = (Trad/Tkin)4 Mλ= εσT4 Incorporating Emissivity (ε) allows calculation of Radiant flux of nonblackbody materials
Death Valley, CA Landsat natural color (left) and TIMS (center
The range in Digital numbers (DNs) is the Radiometric Resolution
What Is A Digital Image? Pixel Digital Number (DN)
70
53
41
64
84
85
81
88
91
87
79
77
45
38
59
77
84
86
85
85
80
82
69
44
32
45
72
86
82
78
88
79
86
87
65
40
41
75
79
78
93
86
93
106
106
84
56
43
58
75
104
104
100
101
95
91
83
51
39
56
105
110
97
88
84
85
87
77
59
44
96
103
89
79
79
75
77
79
74
72
87
93
97
90
82
76
70
67
61
71
79 81
DN are typically ranges from 0 to 255; 0 to 511; 0 to 1023, etc. These ranges are binary scales: 28=256; 29=512; 210=1024.
Current instruments typically have 214 What What your computer you see…sees… or 216 DN = white By convention, 0=black 255 81
88
97
93
85
78
74
70
72
75
78
85
94
97
92
84
80
72
Spectral Bands in Image Dataset Reflected Light from 3 bands
Light Source Each pixel is measured by an individual detector.
B3 B2 B1
Grayscale vs. RGB Grayscale represents the data in the band as an image where the variation in values is represented by the different intensity tones. … RGB (“Red”, “Green”, “Blue”) images are composites of 3 bands, corresponding to the red, green and blue phosphors on a monitor. Computer monitor colors are additive, meaning maximum DN of red + green + blue = white.
