Lecture 18 – TIR remote sensing

Reading assignment“Estimating Sub-pixel Surface Roughness Using Remotely Sensed Stereoscopic Data” Mushkin & Gillespie, pdfpreprint for RSE available on class website

Last lecture – RADAR

Thursday, 5 March

Thermal-Infrared imaging

What is it?

- measurement of emitted radiation (temperature)- at one or more times (thermal inertia) - at one or more wavelengths (composition)

Why bother?

- see at night- temperatures- energy fluxes- material properties (resistance to

temperature change, or thermal inertia)- composition (emissivities)

VNIR-SWIR image of Pu’u O’oErupting(Hawai’i)

Kilauea

Thermal Infrared

•Radiance is measured in the 3-5 or 8-12 µm window

•Generally the goal is to recover temperature

•In mapping applications emissivity may be equally interesting

•Emissivity is like the efficiency with which thermal energy is radiated from a sample

•Emissivity and reflectivity are complementary: ρ=1-ε(Kirchhoff’s “Law”)

Thermal-Infrared imaging

What are the governing equation?

Planck’s Law: R = εε((λλ) ) c1π-1 λλ -5[exp(c2 λλ-1TT-1 )-1] -1

Emissivity

Blackbody radiation

Planck equation - radiance from a blackbody

Bλ =c1

πλ51

exp(c2 / λT( )− 1⎛

⎝ ⎜

⎞

⎠ ⎟

B = blackbody radiance (W m-2 sr-1 µm-1) λ = wavelength (µm)c1 = 2π h c2 (3.74x10-16 W m2; 1st radiation constant) T = temperature (K)h = 6.63x10 -34 W s2 (Planck's constant) c = 2.99x10 8 m s-1 (speed oflight)c2 = h c/k (1.44x104 µm K; 2nd radiation constant) k =1.38x10-23 W s K-1

(Boltzmann's constant)Blackbody radiance in the VNIR-TIR windows

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

0 2 4 6 8 10 12

wavelength, µm

T=300KT=1000Kt=1450KT=5900K

A. R. Gillespie - 19-20 Feb 2003 UW

Near-shore temperatures and energy transfer

AVHRR

ASTER TIR Image

Martha’s Vineyard

8.5 µm 9.1 µm 10.8 µm

UnstretchedDay/Night

62 145

Degrees F

23 July 2001

62 94

Degrees F

1:30 pm

8:00 pm

MorningVeg Mapping - Thermal

Day Night

Red = 10:00 amGreen = 2:00 pmBlue = 11:00 pm

Conifers cooler during daywarmer at night

Emissivity of Lava - July 2002

0.00

0.50

1.00

1.50

2.00

2.50

0 200 400 600 800 1000 1200

Temperature [deg.C]

run1 day1run2 day1run1 day2run2 day2run3 day2run4 day2run5 day2

Emis

sivi

ty

MultispectralMultispectral Thermal imagingThermal imaging

Why bother?

- composition (emissivities)

Planck’s Law

- R(λ) = εε((λλ) ) c1π-1 λλ-5 [exp(c2 λλ-1TT-1 )-1] -1

IN PRINCIPLE IN PRINCIPLE -- if you know T & R you can find if you know T & R you can find εε

- εε((λλ) = ) = c1π-1 λλ-5 [R(λ) exp(c2 λλ-1TT-1) -1]-1

Blackbody radiance

Emissivity & Planck equationThermal radiation

0.38

0.39

0.40

0.41

0.42

0.43

0.44

0.45

0.46

0.47

8 9 10 11 12

wavelength, µm

Blackbodywith emissivity

Emissivity = measured Radiance / BB(T)

Emissivity

0.84

0.86

0.88

0.90

0.92

0.94

0.96

0.98

1.00

8 9 10 11 12

wavelength, µm

Emissivity spectra of rocks

Emissivity Spectra

0.7

0.8

0.9

1.0

8 9 10 11 12 13 14

Wavelength, µm

dolomitelimestonedolomitegypsumvarnish/ss

Emissivity spectra of rocks

Emissivity of rocks

0.5

0.6

0.7

0.8

0.9

1.0

8 9 10 11 12 13 14

wavelength, µm

graniteanorthositebasaltduniteobsidian

Emissivity spectra of approximate graybodies

Emissivity of soil & graybodies

0.80

0.85

0.90

0.95

1.00

8 9 10 11 12 13 14

wavelength, µm

green grassconifersicesnowdry grassspodosol

What compositions can be What compositions can be determined in the TIR?determined in the TIR?

Mostly vibrational resonance, not electronic processestherefore, relatively large molecules

Silicate minerals (SiO4-4); quartz (SiO2)

Sulfates (SO4=); sulfur dioxide (SO2)

Carbonates (CO3=); carbon dioxide (CO2)

Ozone (O3) Water (H2O)Organic molecules

A little about solving sets of equations

If you measure R there are 2 unknowns: ε and TIf you measure R at a different λ, there is another unknown ε

If you measure a spectrum of n bands, there are n+1 unknowns

You must have the same number of measurements as unknowns to solve a set of equations

How can you do this for TIR data?

Temperature - Emissivity Separation

•Two-time two-channel method•Completely determined

•Model emissivity method•Assume ε10µm=0.96

•Normalized Emissivity method•Assume εmax=0.96

•TES algorithm•Assume εmin= a+b[(εmax - εmin) /εavg]c

Death Valley, California

Saline Valley, California

VNIR SWIR TIR

Mauna Loa, Hawaii

MASTER VNIRdaytime

ASTER TIR,daytime

MTI TIR,nighttime

Hyperspectral TIR imaging, Yuma Proving Grounds, Arizona

Next lecture: Roughness from stereo images