52
Principles of Radiative Transfer Principles of Remote Sensing - Marianne König EUMETSAT [email protected]

Principles of Radiative Transfer Principles of Remote SensingATMOSPHERIC VARIABLES T,q(z) RADIATIVE BOUNDARY CONDITIONS RadiativeTransfer Theory: Forward Problem Slide: 37 WSN-12 Rio

  • Upload
    others

  • View
    7

  • Download
    0

Embed Size (px)

Citation preview

Principles of Radiative

Transfer

Principles of Remote

Sensing

-

Marianne KönigEUMETSAT

[email protected]

Remote Sensing

All measurement processes which perform observations/measurements of parameters which carry information about properties at the location of interest, far from the location of interest

Slide: 2 WSN-12 Rio de Janeiro 06-10 August 2012

Opposite: In-situ measurements, i.e. at the location of interest

For Meteorology: most remote sensing relies on electromagnetic waves

Remote Sensing

"Remote Sensing" is for us

not such a strange principle – we have several remote sensing devices (which?)

Slide: 3 WSN-12 Rio de Janeiro 06-10 August 2012

Remote Sensing - Principle

In order to obtain meaningful information from remote sensing images, the probed radiation field must have some interaction with the parameter of interest

Example:

A fish seen in "visible"

Slide: 4 WSN-12 Rio de Janeiro 06-10 August 2012

A fish seen in "visible" wavelengths (by humans)

Remote Sensing - Principle

In order to obtain meaningful information from remote sensing images, the probed radiation field must have some interaction with the parameter of interest

Example:

A fish seen in "visible" And seen by x-rays

Slide: 5 WSN-12 Rio de Janeiro 06-10 August 2012

A fish seen in "visible" wavelengths (by humans)

And seen by x-rays

Remote Sensing - Principle

Slide: 6 WSN-12 Rio de Janeiro 06-10 August 2012

Another example: photograph and infrared picture of a house

Electromagnetic Waves

Slide: 7 WSN-12 Rio de Janeiro 06-10 August 2012

Characteristics:• Wavelength λ• Propagation velocity c • Frequency ν = c / λ• Wavenumber =1/ λ (cm-1)

Electromagnetic Spectrum

1m 1mm 1µm

Slide: 8 WSN-12 Rio de Janeiro 06-10 August 2012

Sources of Radiation (Met Applications)

Slide: 9 WSN-12 Rio de Janeiro 06-10 August 2012

Electromagnetic Radiation – Units and Concepts

Irradiance Watts/meter2

Total energy which falls onto 1 sqm of surface

Slide: 10 WSN-12 Rio de Janeiro 06-10 August 2012

Electromagnetic Radiation – Units and Concepts

Radiance Watts/meter2/ster (W/m2/ster)

Energy which falls onto 1 sqm of surface, coming from a certain direction

Can also be expressed as radiance per wavelength or wavenumber:

W/m2/ster/µm

Slide: 11 WSN-12 Rio de Janeiro 06-10 August 2012

W/m /ster/µm

W/m2/ster/cm-1

Fundamental Radiation Law: Planck’s Law

Slide: 12 WSN-12 Rio de Janeiro 06-10 August 2012

1

12),(

2

3

−=

kThec

hTB

ν

νν1

1),(

5

2

−=

kThce

hcTB

λλλ

k = Boltzmann‘s constant

T = Temperatureh = Planck‘s constant

Fundamental Radiation Law: Planck’s Law

Slide: 13 WSN-12 Rio de Janeiro 06-10 August 2012

Spectral Distribution of Energy Radiated

from Blackbodies at Various Temperatures

Slide: 14 WSN-12 Rio de Janeiro 06-10 August 2012

P. Menzel, 2007

Concept of a Blackbody – Concept of Emissivity

A “Blackbody” is an object of temperature T which radiates energy according to Planck’s Law. Nature does not have perfect blackbodies:

Slide: 15 WSN-12 Rio de Janeiro 06-10 August 2012

Satellite Orbits – What Can We Measure?

Geostationary orbit:36000 km heightUsable energy in solar and infrared bands

Low earth / polar orbit:~800-900 km heightUsable energy in solar, infrared and

Slide: 16 WSN-12 Rio de Janeiro 06-10 August 2012

Usable energy in solar, infrared and microwave bands

Remote Sensing of the Atmosphere

What do we measure?

Solar radiation: reflected by the surface, by clouds, scattered by molecules … (wavelengths?)

Slide: 17 WSN-12 Rio de Janeiro 06-10 August 2012

Thermal radiation: emitted by the earth / clouds / atmosphere …

(wavelengths?)

What about thermal radiation from the sun???

Explanation

At our Earth’s distance from the sun, the radiation received from the sun is approximately on the same energy level as the radiation emitted from the earth/atmosphere

Slide: 18 WSN-12 Rio de Janeiro 06-10 August 2012

Visible(Reflective Bands)

Infrared / microwave(Emissive Bands)

microwave

Slide: 19 WSN-12 Rio de Janeiro 06-10 August 2012

P. Menzel, 2007

Processes for Solar Radiation

Slide: 20 WSN-12 Rio de Janeiro 06-10 August 2012

Why is grass green?

Processes of Thermal Radiation

Slide: 21 WSN-12 Rio de Janeiro 06-10 August 2012

Q

O3

Earth Spectrum and Planck Curves

Slide: 22 WSN-12 Rio de Janeiro 06-10 August 2012

CO2

H20

CO2

Radiative Processes

Absorption: Energy of the electromagnetic wave is taken up by matter (e.g. change in atomic state)

Emission: Energy change in the matter (e.g. change in the atomic state) releases electromagnetic radiation

Emission = Absorption!!

Slide: 23 WSN-12 Rio de Janeiro 06-10 August 2012

Emission = Absorption!!

Absorption coefficient = property of matter

Radiative Processes

Absorption: Energy of the electromagnetic wave is taken up by matter (e.g. change in atomic state)

Emission: Energy change in the matter (e.g. change in the atomic state) releases electromagnetic radiation

Emission = Absorption!!

Slide: 24 WSN-12 Rio de Janeiro 06-10 August 2012

Emission = Absorption!!

Absorption coefficient = property of matter

Scattering/reflection: Radiation is “geometrically” forced to deviate from a straight line

Illustration: Beam at 11 µm wavelength (“Window”)

Sensor

Slide: 25 WSN-12 Rio de Janeiro 06-10 August 2012

Earth Surface

Temperature Profile

Illustration: Beam at 6.5 µm wavelength (WV Absorption)

Sensor

Slide: 26 WSN-12 Rio de Janeiro 06-10 August 2012

Earth Surface

Temperature Profile

Weighting Functions

height

Absorption Channel:

peaks high in the

atmosphere

Slide: 27 WSN-12 Rio de Janeiro 06-10 August 2012

0 1

atmosphere

Window Channel: High

contribution from

surface

Question: What happens for higher viewing angles?

A) Satellite measures the same brightness temperaturesB) Satellite measures higher brightness temperaturesC) Satellite measures lower brightness temperatures

Sensor

Slide: 28 WSN-12 Rio de Janeiro 06-10 August 2012

Earth Surface

Temperature Profile

Question: What happens for strong absorption channels,

at higher viewing angles?

A) Satellite measures the same brightness temperaturesB) Satellite measures warmer brightness temperaturesC) Satellite measures colder brightness temperatures

Sensor Q

Slide: 29 WSN-12 Rio de Janeiro 06-10 August 2012

Earth Surface

Temperature Profile

Radiative Processes Can Be Modelled - RTMs

The equation of radiative transfer simply says that as a beam of radiation travels, it loses energy to absorption, gains energy by emission, and redistributes energy by scattering.

Slide: 30 WSN-12 Rio de Janeiro 06-10 August 2012

The equation is a differential equation, numerical models exist which provide a solution (Radiative Transfer Models, RTMs)

Practical Example: MODIS Imagery, 03 April 2011

Solar Bands

0.6 µm 0.9 µm 1.6 µm

Practical Example: MODIS Imagery, 03 April 2011

Thermal Bands

266.9 K 237.6 K 249.3 K

11 µm 13.2 µm 7.3 µm

218.5 K 218.8 K 220.9 K

Scattering

- Scattering by particles which are much smaller than the electromagnetic wavelength ("Rayleigh Scattering")

- Scattering by particles which are of same size and larger than the electromagnetic wavelength ("Mie Scattering")

Slide: 33 WSN-12 Rio de Janeiro 06-10 August 2012

Distribution for all angles: phase function

Rayleigh Scattering

Rayleigh scattering, named after the British physicist Lord Rayleigh, is the elastic scattering of light or other electromagnetic radiation by particles much smaller than the wavelength of the light. The particles may be individual atoms or molecules.

Slide: 34 WSN-12 Rio de Janeiro 06-10 August 2012

or molecules.

Scattering is ~ λ-4, i.e. scattering occurs for shorter wavelengths!

Water Clouds: Scattering on Spherical Particles

Size distribution Wavelength 0.87 µmCloud droplets 1- 10 µm

Strong forward scattering

Slide: 35 WSN-12 Rio de Janeiro 06-10 August 2012

Ice Clouds: Complex Scattering Depending on Ice

Crystals' Shape

Plates

Columns

Slide: 36 WSN-12 Rio de Janeiro 06-10 August 2012

Columns

Rosettes

Aggregates

ATMOSPHERICVARIABLES

T,q(z)

RADIATIVE

BOUNDARYCONDITIONS

Radiative Transfer Theory: Forward Problem

Slide: 37 WSN-12 Rio de Janeiro 06-10 August 2012

HydrometeorsLiquid waterCloud type

etc.

RADIATIVETRANSFEREQUATION

RADIOMETERCHARACTERISTICS

TB

ATMOSPHERICVARIABLES

T,q(z)

RADIATIVE

BOUNDARYCONDITIONS

Radiative Transfer Theory: Inverse Problem

Slide: 38 WSN-12 Rio de Janeiro 06-10 August 2012

HydrometeorsLiquid waterCloud type

etc.

RADIATIVETRANSFEREQUATION

RADIOMETERCHARACTERISTICS

TB

Retrieval/InversionScheme

T(p)

q(p)

p

T(p)

q(p)

p

T(p)

q(p)

p

T(p)p

T(p)

q(p)

p

TBs in different wavelengths

Inversion Problem

Slide: 39 WSN-12 Rio de Janeiro 06-10 August 2012

ILL POSED PROBLEM

T(p)

q(p)

p

T(p)

q(p)

p

T(p)

q(p)

p q(p)

...Many possible states ofTemperatureWater vapour, etc.(or cloud parameters, aerosol information ….)

Retrieval/InversionScheme

Many TBs in many different wavelengths

Inversion Problem

T(p)

q(p)

p

T(p)

q(p)

p

T(p)

q(p)

p

T(p)p

T(p)

q(p)

p

Slide: 40 WSN-12 Rio de Janeiro 06-10 August 2012

More channels = more information!

T(p)

q(p)

p

T(p)

q(p)

p

T(p)

q(p)

p q(p)

...

Inversion Problem: Practical Example, 11µm

Satellite Measurement: 286 K

286 K 286 K286 K

RTM result

Slide: 41 WSN-12 Rio de Janeiro 06-10 August 2012

291 Kε = 0.99

Some more H2O

292 Kε = 0.99

Even more H2O

290 Kε = 0.99

Little H2O

Satellite Measurement: 286 K

286 K 286 K286 K

RTM result

Inversion Problem: Practical Example, 11µm

Q

Slide: 42 WSN-12 Rio de Janeiro 06-10 August 2012

291 Kε = 0.99

Some more H2O

292 Kε = 0.99

Even more H2O

290 Kε = 0.99

Little H2O

Which is the correct surface temperature?

Can we tell from this one measurement?

Satellite Measurement: 286 K

286 K 286 K286 K

RTM result

Inversion Problem: Practical Example, 11µm

Slide: 43 WSN-12 Rio de Janeiro 06-10 August 2012

291 Kε = 0.99

Some more H2O

292 Kε = 0.99

Even more H2O

290 Kε = 0.99

Little H2O

No, we cannot tell!

Possible: constrain humidity by forecast profile

Or: combine with another channel that is sensitive to surface temperature and humidity

HIRS Ch01

ca. 23 km

Slide: 44 WSN-12 Rio de Janeiro 06-10 August 2012

HIRS Ch02

ca. 19 km

Slide: 45 WSN-12 Rio de Janeiro 06-10 August 2012

HIRS Ch03

ca. 17 km

Slide: 46 WSN-12 Rio de Janeiro 06-10 August 2012

HIRS Ch04

ca. 7 km

Slide: 47 WSN-12 Rio de Janeiro 06-10 August 2012

HIRS Ch05

ca. 4 km

Slide: 48 WSN-12 Rio de Janeiro 06-10 August 2012

HIRS Ch06

ca. 2 km

Slide: 49 WSN-12 Rio de Janeiro 06-10 August 2012

Outlook: Hyperspectral Measurements

SurfaceCloudsSurface

CloudsTemp(CO2)

SurfaceClouds

Instruments like IASI measure the IR spectrum in 8461 different samples

Slide: 50 WSN-12 Rio de Janeiro 06-10 August 2012

Temp(CO2)

O3

H2O,CH4,N2O

CO

N2O,Temp(CO2)

(CO2)

IASI Example

Slide: 51 WSN-12 Rio de Janeiro 06-10 August 2012

The End

Thank you for your attention!

Consider yourself "remote sensing experts" now!

Slide: 52 WSN-12 Rio de Janeiro 06-10 August 2012