passive microwave observationsjoseba.mpch-mainz.mpg.de/pdf_dateien/microwave.pdf · spectral lines,...

Post on 06-Aug-2020

6 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Passive microwave observations

Why?

several advantages:

-pressure broadening can yield profile informatiom

-usually, microwave observations are not affected by clouds and rain, because the wavelenghts are much larger than the droplet sizes

The term ‘Microwave’ is typically used to denote measurements at centimeter, millimeter and submillimeter wavelengths.

λ = 0.1 – 10 cm ν ~ 3 -300GHz (3 THz)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Passive microwave observations

Measurment modes:

-typically emissions are measured (day and night)

-also absorption measurements are possible (e.g. Sun light)

-spectroscopic measurements allow the observation of atmospheric trace gases (and other parameters like temperature from rotational transitions) => microwave sounders (atmosphere is the light source)

-broad band measurements can yield information on surface properties

=> microwave imagers (surface is the light source) also atmospheric properties (e.g. humidity, rain rate, etc. can be retrieved)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

0.0E+00

1.0E-12

2.0E-12

3.0E-12

4.0E-12

5.0E-12

6.0E-12

0 5000 10000 15000 20000 25000 30000 35000 40000

Frequency [GHz]

B(

)d [

W/m

²]

50K100K200K300K

Microwave region

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

0.0E+00

5.0E-15

1.0E-14

1.5E-14

2.0E-14

2.5E-14

3.0E-14

3.5E-14

4.0E-14

0 100 200 300 400 500 600 700 800 900 1000

Frequency [GHz]

B(

)d [W

/m²]

50K100K200K300K

kThe kT

h νν

−≈−

1with 1<<kThν

=>

=> ( ) 2

22c

kTB νν ≈Rayleigh Jeans radiation law

-linear in T

-square dependence in ν

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

1.0E-22

1.0E-21

1.0E-20

1.0E-19

1.0E-18

1.0E-17

1.0E-16

1.0E-15

1.0E-14

1.0E-13

0.1 1.0 10.0 100.0 1000.0

Frequency [GHz]

B(ν)

d ν [W

/m²]

50K100K200K300K

UH

F (TV)

VH

F (TV)

Microw

ave oven

FM (R

adio)

Mobile P

honesTurbulent E

ddie profiler

Moisture R

adiometer

Airborne cloud R

adar

MLS

ClO

observations

AS

UR

O3, C

lO, etc.

3m 30cm 3cm 3mm 0.3mm

SS

M/I, surface

properties

OH

Radiom

eter

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

General rule:

Molecular signatures aremeasured at high

frequencies

Surface properties are measured at low

frequencies

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Graybody and Selective Emitters

• Graybody emitters are those whose emittance is less than a black body with the same temperature, but whose emissions– are constant for any wavelength– are a consistent fraction of the blackbody emittance

• Selective emitters are bodies whose emittance is less than a black body with the same temperature, but not constant as a function of wavelength

Brightness Temperature

For a surface with an emissivity, e, the brightness temperature becomesTB = e TG

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The observed power as function of viewing angle is:

If only one polarisation direction is measured

Intensity received from direction Θ, Φ

Effective antenna area for directionΘ, Φ

© Joe Waters, Microwave (Ground, Aircraft, Balloon, Satellite) Observations for Chemistry (Remote Sensing)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The observed power as function of viewing angle is:

If only one polarisation direction is measured

Intensity received from direction Θ, Φ

Effective antenna area for directionΘ, Φ

= 0 if only emission is observed Optical depth

In the microwave spectral region (and also in the thermal IR) typically emission and absorption have to be considered

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Diffraction limits the received power:

-for microwave observations, typically the aperture angle of the telescope is determined by diffraction

-it is approximately inverse proportional to the collecting area

-thus the effective collective area (integrated over all viewing angles) for isotropic radiation is independent on the size of the antenna.

∫ Aeff = λ²

Small telescope, large aperture

large telescope,small aperture

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Diffraction limits the received power:

∫ Aeff = λ²

-for temperatures in the Earth’s atmosphere ≈ 1 / λ²

Thus λ² B(T) is nearly constant with frequency up to ~5 THzBeyond it drops rapidly. Microwave measurements of atmospheric thermal emission are, thus, relatively efficient at frequencies up to ~5 THz ( λ >~6*10-5 m) but not much beyond.

( ) 2

22c

kTB νν ≈

0.0E+00

1.0E-12

2.0E-12

3.0E-12

4.0E-12

5.0E-12

0 5000 10000 15000 20000 25000 30000 35000 40000

Frequency [GHz]

B(

6.0E-12

)d [

W/m

²]

50K100K200K300K

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The power per unit frequency interval at frequency ν from a blackbody radiation field at 200 and 250 K (plotted in units of W / MHz, as MHz is the typical unit for widths of microwave spectral lines).

(From Waters, J.W. (1993) Microwave Limb Sounding. chapter 8 in Atmospheric Remote Sensing by Microwave Radiometry. M.A. Janssen, ed. New York: John Wiley. 1993 Wiley)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The observed power as function of viewing angle is:

If only one polarisation direction is measured

Intensity received from direction Θ, Φ

Effective antenna area for directionΘ, Φ

The observed power as function of wavelength and temperature:

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Atmospheric zenith transmission above a 4 km high mountain (top) and abovean aircraft at 12 km (bottom). The absorption features seen here are due to H2O and, to a lesser extent, O2.

(From Phillips, T. G., and Keene, J. (1992) Submillimeter Astronomy, Proceedings of the IEEE 80, pp 1662-1678. � 1992 Institute of Electrical and Electronics Engineers, Inc.)

0.1 mm1 mm

Atmospheric transparency

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

278 GHz

The atmospheric transparency depends strongly on the atmospheric water vapor content

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

( ) ( ) ),,(,,,, TpSTNNTp ijijij νννσνα −⋅⋅=

N: Number densityT: Temperaturep: Pressureσ: absorption cross sectionS: Line width

α: absorption coefficient

gl is the degeneracy and El the energy for state l,μ is the overall dipole (or other) moment coupling to the radiation fieldQ(T)= Σgl exp (- El /kT) is the partition functionφij² is the transition matrix element

Optical depth:

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Natural line width:

It can usually be ignored compared to collision (pressure) broadening at lower altitudes and Doppler (thermal motion) broadening at higher altitudes.

Collision (pressure) broadening:

-collission between molecules shortens lifetimes for specific states-for increasing pressure towards lower altitudes the probability of collisions increases-the line shape for pressure broadening can be approximated by a Lorentzianline shape:

(Δν: line width for pressure broadening)

The Lorentian line shpae can be retrievd from the ‚van Vleck and Weisskopf‘ line shape assuming that the duration of a collision is much shorter than the time between two collisions (impact-approximation)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lorentian line width:

The Lorentian line width is temperature dependent

Typical value: 2.5 MHz Typical value: 0.75p0: 1hPaT0: 300K

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Doppler broadening

-is caused by thermal motion of molecules. The Maxwell distribution depends on temperature and the molecular mass:

According to the Doppler-effect, the line fuction becomes:

With line width:

By convolution of the Lorentian and Doppler line shape one gets the so called Voigt line shape:

Heavy molecules have higher velocities

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Line width (half width at half maximum) versus altitude for some representative microwave spectral lines.

collision line widths generally increase with dipole moment; O2 has a very small (magnetic) one

Heavy molecules have higher velocities

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Microwave emission lines for the same mixing ratio of a gas at the bottom (100 hPa, ~15 km), middle (10 hPa, ~30 km), and top (1 hPa, ~50 km) of the stratosphere.

For trace gases close to the surface (1000hPa), the measurements often become insensitive

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Where can spectral parameters be obtained?

-Line frequencies are typically known to ~0.1 MHz or better; strengths to ~1% or better.

-Dipole (and other) moments are measured from Stark and Zeeman splitting of spectral lines, allowing line strengths to be determined without requiring measurement of the gas abundance in a laboratory cell.

-Line width parameters can be measured with ~3% accuracy.

Data bases of these parameters exist and are continuously updated (e.g. catalog maintained by the Molecular Spectroscopy Group at the California Institute of Technology Jet Propulsion Laboratory (see world wide web site http://spec.jpl.nasa.gov)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Spectra of some stratospheric molecules and atomic oxygen. Vertical axis is thelogarithm of the optical depth for an observation path through the atmospheric limb with the indicated volume mixing ratios (vmr).

(From Waters, J.W. (1993) Microwave Limb Sounding, chapter 8 in Atmospheric Remote Sensing by Microwave Radiometry. Janssen, M.A. ed. New York: John Wiley, 1993 Wiley)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Spectrometers and detectors

In the microwave region, the atmospheric emission is measured in small spectral channels using so-called radiometers or heterodyne techniques.

Radiometers

Passive microwave imaging radiometers (usually called microwave imagers) collect Earth's natural radiation with an antenna and typically focus it onto one or more feed horns that are sensitive to particular frequencies and polarizations. From there, it is detected as an electrical signal, amplified, digitized.

The observations are sensitive to environmental parameters such as soil moisture content, precipitation, sea-surface wind speed, sea-surface temperature, snow cover and water content, sea ice cover, atmospheric water content, and cloud water content.Observations are typically made in zenith geometry.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Spectrometers and detectors

In the microwave region, the atmospheric emission is measured in small spectral channels using so-called radiometers or heterodyne techniques.

heterodyne techniques - spectrometers

‘Heterodyne’ indicates multiplying a weak input signal by a strong local oscillator signal to translate - without loss of information - the input signal to a portion of the spectrum more convenient for further processing. This process allows measurements of weaker signals, and better spectral resolution, than generallycan be obtained with other techniques. Technology for low-noise submillimeter heterodyne measurements has become available only recently, and is advancing rapidly.

The observations are sensitive to atmospheric trace gas emissions, especially from the stratosphere. Observations are typically made in limb geometry.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

heterodyne techniques - spectrometers

-Atmospheric signals are collected by an antenna, possibly amplified or filtered, and passed to a mixer.

-The mixer combines the atmospheric signal with a monochromatic local oscillator (LO) signal in a non-linear process that reproduces the signal spectra at frequencies that are sums and differences of the atmospheric and LO signal frequencies, and possibly at their harmonics.

-The LO frequency is chosen so the mixing product of interest (usually at the difference between the LO and signal frequencies) appears at intermediate frequencies (IF) convenient for further processing.

-The IF signal, after amplification, is passed to a spectrometer that separates it into individual spectral channels with desired resolution. Each channel’s signal is then ‘detected’ (converted to a voltage proportional to its power), digitized and passed to the data handling system.

-Instruments often contain multiple mixers, LOs, IF amplifiers and spectrometers to allow simultaneous observations of several spectral lines.

-LO frequencies can be changed operationally to change measurements.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Typical block diagram of an instrument for microwave observations of stratosphericchemistry. The ‘radio frequency’ amplifier is currently available only at lower frequencies and does not appear in many systems; a filter is sometimes placed at this position to eliminate unwanted signals in one of the mixer’s sidebands. The portion of the instrument between the antenna and spectrometer is called the ‘receiver’ or ‘radiometer’.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Microwave instruments of different complexity

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

State-of-the-art mixers are based on planar Schottky-diodes (either cooled or room temperature), superconductor-insulator-superconductor (SIS) tunnel junctions, or superconducting hot electron bolometer (HEB) devices.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The I-V dependence of a diode has an exponential form,

where and h are constants whose values depend upon the details of the diode, T is the diode's temperaturein Kelvins and k is Boltsmann's constant. In practice, the shape of a diode's I-V curve depends on the details of how it was made, hence many real diodes only approximate to exponential behaviour.

http://www.st-andrews.ac.uk/~www_pa/Scots_Guide/RadCom/part1/page1.html

For our purposes it's convenient to assume that we have a Square Law diode whose behaviour obeys

many real diodes are deliberately manufactured to behave in a square law manner, so the assumption turns out to be a good description of reality. Let‘s apply a combination of three input voltages to a square law diode:

Atmospheric signal oscillator signal

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The signal of the difference frequency fs- fL can be easiliy

filtered out and further processed

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Total signal: atmospheric emissions and noise

Kleinböhl, 2003:

Calibration is performed by inserting targets (typically blackbody targets - with ‘cold space’ generally used for one target when possible) in the signal path near the

instrument input, ideally before the antenna.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The sensitivity of a microwave radiometer is usually specified by its ‘receiver noisetemperature’. The root mean square measurement noise, expressed in temperature units, for integration time ∆t and spectral resolution ∆ν is given by

where Trec is the receiver noise temperature, and a ≈ 1 for a non-chopped system or a ≈ 2 for a chopped system. Tsig, usually much smaller than Trec for thermal emission measurements, is present because the signal itself is noisy.

-The value of Trec is primarily set by noise in the first amplifier, typical values are~400 K for cooled radiometers and 2000K for room temperature radiometers at ~200 GHz. -The measurement noise with ∆t=1 s integration, for example, is ∆Trms ≈ 1 K for Trec ≈ 4000 K and ∆Trms ≈ 0.1 K for Trec ≈ 400 K. -Increasing the integration time to 1 hour gives ∆Trms ≈ 0.02 K for Trec ≈ 4000 K and ∆Trms ≈ 0.002 K for Trec ≈ 400 K.

Microwave radiometer noise generally increases with frequency, but this can be offset by increases in line strength. Improvements are expected as this technology matures.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Several types of spectrometers are currently used:

-Filter banks have a set of simultaneously observed spectral filters, usually made of discrete or distributed capacitative and inductive elements, whose frequencies and widths are set as needed for a particular measurement.

-Digital autocorrelators measure a signal’s autocorrelation simultaneously at many time lags; Fourier transforming the measured autocorrelation function gives the spectrum.

-Acousto-optic spectrometers use the IF signal to modulate a Bragg cell which, according to the signal’s spectral content, diffracts a laser beam to a detector array.

-Chirp transform spectrometers multiply the IF signal by a frequency-modulated (chirp) waveform, convolve the resulting product with a filter that is appropriately matched to the chirp, and the spectrum appears as a function of time at the output.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Measured (crosses, bars) and calculated (thin line) stratospheric emission spectrum in bands measured simultaneously near 626 and 649 GHz. ‘USB’: upper sideband, ‘LSB’: lower sideband, ‘IF’: intermediate frequency. LSB spectral lines are in sans serif font; USB lines are in italicized roman font

(Adapted from Stachnik, R.A., et al. (1992) Submillimeterwave Heterodyne Measurements of Stratospheric ClO, HCl, O3 and HO2: First Results. Geophys. Res. Lett. 19, 1931-1934.)

Satellite microwave observations made

at discrete channels

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Acusto-optical spectrometer (AOS)

-IF signal is transformed in ultrasonic signal

-ultrasonic signal is connected to transparent crystal

-the sound waves cause density fluctuations with different wavelength and amplitude

-laser illuminates the crystal homogeneously

-the density fluctuations act like a difractive grating

-the microwave signal is transformed into a ‚optical spectrum‘ abd is read out using a CCD array

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Acousto optical spectrometer:

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Ground based observations

-can provide continuous monitoring at selected sites

Vertical resolution is obtained from the spectral line shape, it is typically around one atmospheric pressure scale height (~6-8 km)

-Initial ground-based microwave measurements in the 1970s: -stratospheric and mesospheric O3 from lines near 100 GHz-stratospheric temperaturesand from rotational lines of O2 (60 GHz)

-Ground-based microwave instruments are currently used by several research groups, and are deployed in the international Network for the Detection of Stratospheric Change to measure stratospheric ClO, O3 and H2O.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Ground based observations of stratospheric ClO

Ground-based microwave measurements have provided important results forunderstanding stratospheric chlorine chemistry.

-In 1981 they gave the first definitive remote measurements of ClO, the key chlorine radical involved in ozone depletion. Early results also included the first measurement of ClO diurnal variation, testing crucial aspects of upper stratospheric chlorine chemistry.

-In 1986 the technique gave the first evidence of greatly enhanced ClO in the Antarctic lower stratosphere, firmly connecting chlorine chemistry to the ozone hole.

-Measurements of ClO diurnal variation tested chemical models for formation and photolysis of the ClO dimer in the Antarctic lower stratosphere.Enhanced ClO in the Arctic winter stratosphere also has been measured.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

De Zafra et al, Nature, 1987

278.631 GHz

First observation of ClO

a) Raw spectrumb) O3-Spectrumc) Differenced) Nightime spectrum (no ClO)e) Nightime spectrum (d) subtracted from Daytime spectrum (c)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

De Zafra et al, Nature, 1987

2-h-IntervalsAverages of 22 days

Increase of broad wings of line→ClO at low altitudes

Diurnal variation of ClO

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Ground-based 278 GHz measurement of stratospheric ClO over Antarctica. The left panel shows day-night differences of the spectral line measured during several days in 1992. The right panel shows a height-time cross section of the retrieved ClO mixing ratio profile, where

contours are in parts per billion by volume. (From deZafra, R.L., Reeves, J.M., and Shindell, D.T. (1995) Chlorine monoxide in the Antarctic spring vortex 1. Evolution of

midday vertical profiles over McMurdo Station, 1993. J. Geophys. Res. 100, pp 13,999-14,007. � 1995 American Geophysical Union)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The radiometer MIAWARA on the roof of theInstitute of Applied Physics at Bern, Switzerland

(46.95 N / 7.45 E, 550 m. above sea level).

Measured balanced spectrum from April 2002

The measured spectra (blue: narrowband spectrometer;green: broadband spectrometer; integration-time: 1.4 hours) are corrected for tropospheric attenuation. Thered line shows a spectrum calculated from a standard

water wapor profile.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Spectra of O3 and H2O18 measured with EMCOR from Jungfraujoch with corresponding synthetic spectrum based on the retrieved profile

Comparison of vmr for H2O18 deduced from EMCOR with data from ATMOS, an IR solar occultation experiment, and data from a balloon borne Fourier transform instrument

N.Kämpfer Institute of Applied Physics, Univ. of Bern, Sidlerstrasse 5, CH-3012 Bern, Switzer-land

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Aircraft-based Observations

-good horizontal resolution along a measurement track over an extended spatial range.

-Instruments can observe in high-frequency spectral windows where troposphericH2O absorption prevents ground-based measurements and where more species have spectral lines.

-Vertical resolution is obtained from spectral line shape for measurements above the aircraft altitude, and can be obtained from limb sounding techniques at heights below the aircraft.

-Initial aircraft measurements in the 1970s included stratospheric H2O and O3 from lines near 183 GHz, and an upper limit on stratospheric ClO abundance.

-Recent measurements include stratospheric HCl, ClO, O3, HNO3, N2O, H2O, HO2, BrO and volcanic SO2

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Aircraft-based Observations

Aircraft measurements of stratospheric HCl (triangles, from the 626 GHz line) and ClO(diamonds, from the 649 GHz line) from a January 2000 flight through the edge of the Arctic vortex. These results show the transition of stratospheric chlorine from the relatively-inert HCl at lower latitudes to the highly-reactive ClO at higher latitudes inside the vortex.

Bremer, H., et al. (2002) Ozone depletion observed by ASUR during the Arctic Winter 1999/2000. J. Geophys. Res. 107

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.deOffice OBR - Atmospheric radiative transfer - Microwave radiometry.htm

geometry for surface emissivityretrieval

location of microwave radiometers on the BAe 146-301.

MIAWARA - Middle Atmospheric Water Vapour Radiometer

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Balloon-based microwave observations

-can provide measurements throughout the stratosphere with 2-3 km vertical resolution.

-The instrument FOV is vertically-scanned through the atmospheric limb to observe a long path length and to obtain the vertical resolution.

-Balloon instruments provide measurements to higher altitudes with better resolution than can be obtained from aircraft or ground

-they provide valuable development and tests of techniques to be deployed on satellites.

-Initial measurements in the 1980s were of ClO and O3 from lines near 205 GHz.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Balloon microwave measurements of ClO, HCl and HO2. Top panels show thespectral lines for a limb observation path through the middle stratosphere. The bottom panels show retrieved mixing ratio profiles (thick) and uncertainty limits (thin) (Adapted from Stachnik, R.A., et al. (1992) Submillimeterwave Heterodyne Measurements of Stratospheric ClO, HCl, O3 and HO2: First Results. Geophys. Res. Lett. 19, 1931-1934)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Satellite-based Observations

-can provide global coverage on a daily basis.

-Limb-sounding is used for chemistry observations because of its vertical resolution and long path length for observations of small concentrations.

-Low-orbit (~700 km altitude) satellites have an observation path tangent point ~3000 km from the instrument.

-Vertical resolution of ~3 km, for example, then requires an antenna having vertical dimension of ~1000 wavelengths.

-Optically-thin O2 emission and spectral line shapes provide information on the atmospheric pressure level at which the instrument field-of-view is pointed.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Satellite-based Observations

-The need for global measurements of ClO motivated initial development of a satellite microwave instrument for stratospheric chemistry.

-The Microwave Limb Sounder (MLS) - operating unchopped in bands around 63, 183 and 205 GHz - was developed to measure ClO, O3 and H2O from NASA’s Upper Atmosphere Research Satellite (UARS) launched in 1991.

-A somewhat similar instrument, the Millimeter-wave Atmospheric Sounder (MAS) was flown on 3 Space Shuttle missions between 1992 and 1994.

-MLS results showed enhanced ClO filling the lower stratosphere polar winter vortices of both the Antarctic and Arctic. This finding was especially important for the Arctic, demonstrating the effectiveness of localized PSCs in activating chlorine throughout the vortex.

-HNO3 has been measured, providing insights into PSC microphysical processes and quantifying denitrification (removal of nitrogen) differences between the Arctic and Antarctic.

MLS H2O measurements led to discovery of the atmospheric ‘tropical tape recorder’ whereby H2O entering the tropical stratosphere is imprinted with a signature (the corresponding H2O saturation mixing ratio) of the seasonally-varying tropopause temperature.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The UARS MLS Instrument

The figures above show a photo and sketch of UARS MLS. The instrument has three assemblies: sensor, spectrometer and power supply. Thermal control of the sensor is radiational by louvers, with in-orbit temperature stability of approximately0.01oC or better, allowing "total power" measurements which do not require fast switching to a reference. The overall instrument mass is 280 kg, power consumption is 163 W fully-on, and data rate is 1250 bits/second.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Fig. 4. Chemical species measured by individual EOS MLS radiometers. (All radiometers contribute to measurement of temperature and cloud ice.) Different colors indicate different chemical families. Solid lines indicate generally useful precision for individual profiles. Dotted lines indicate that zonal (or other) aver-ages are generally needed to obtain useful precision. Solid lines for CH3CN and 190 GHz ClO are for enhanced abundances (biomass burning injections of CH3CN, and polar winter vortex ClO).

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Earth’s lower stratosphere in the Northern Hemisphere on 20 February 1996 (top)and in the Southern Hemisphere on 30 August 1996 (bottom). White contours show the dynamical edge of the polar vortices. HNO3, ClO and O3 are from MLS. Temperature data are from U.S. National Center for Environmental Prediction.(From Waters J.W., et al. (1999) The UARS and EOS Microwave Limb Sounder Experiments. J. Atmos. Sciences 56, pp 194-218. �

1999 American Meteorological Society)

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

The weak 204 GHz H2O2 line from satellite measurements made over a period of 38 days, (~2 days averaging time). Horizontal bars give the spectral resolution of individual filters and vertical bars give the ∆Trms measurement uncertainty. The 0.353 K background is emission from the lower atmosphere received through the antenna sidelobes.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Temperature Weighting Functions of Saturated MLS RadiancesPlotted in the following are MLS temperature weighting functions at 18km tangent height as a function of only height/pressure, where horizontal differences of the weighting functions are ignored. The MLS 63GHz radiometer has 15 spectral

channels distributed symmetrically around the O2 lines. The channels farther away from the line center have less absorption and therefore provide information at lower

altitudes. The layer thickness is ~10km for channels 1-7, 9-15, and ~15km for channel 8 that is at the line center.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Since 2003: The Earth Observing System (EOS) MLS.

It is an improved version compared to the UARS MLS in providing 1) more and better upper-tropospheric and lower-stratospheric measurements, 2) additional stratospheric measurements for chemical composition and long-lived dynamical tracers, 3) better global coverage and spatial resolution, and 4) better precision for many measurements.

These improvements are possible because of 1) advances in microwave technology that provide measurements to higher frequencies where more molecules have spectral lines and spectral lines are stronger, and provide greater instantaneous spectral bandwidth for measurements at lower altitudes; 2) a better understanding of the capabilities of the measurement technique as a result of the UARS experience; and 3) the EOS near-polar (98° inclination, sun synchronous) orbit that allows nearly pole-to-pole coverage on each orbit, whereas the UARS orbit (57° inclination) and its precession forces MLS to switch between northern and southern high-latitude measurements on an approximate monthly basis

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Illustrating coverage of the EOS MLS ‘standard’ 25-channel spectrometer. Each filter in the spectrometer is shown as a horizontal bar whose width gives the filter resolution. Vertical bars, too small to be seen except in the narrow center channels, give the ±1σ noise for the 1/6 s individual integration time of the MLS instrument and a double-sideband radiometer noise temperature of 1000 K. The signal shown here is a simulated ozone line for a limb observation path with tangent height in the lower stratosphere. The 11 filters between the dotted vertical lines also occur in the ‘mid-band’ spec-trometers.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Examples of target spectral lines measured by five different MLS radiometers and ‘standard’ 25-channel spectrometers. The target line, centered in the band, is labeled in each panel. Additional lines of O3 evident in some bands, and the vibrationally-excited O2 line in band 1 are indicated. The two fine structure components for OH are easily seen, and the three for HCl can be seen with more scrutiny. These are EOS MLS ‘first light’ measurements: OH on 24 July 2004, the others on 27 July.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Spectrometers and detectors

In the microwave region, the atmospheric emission is measured in small spectral channels using so-called radiometers or heterodyne techniques.

RadiometersPassive microwave imaging radiometers (usually called microwave imagers) collect Earth's natural radiation with an antenna and typically focus it onto one or more feed horns that are sensitive to particular frequencies and polarizations. From there, it is detected as an electrical signal, amplified, digitized. The observations are sensitive to environmental parameters such as soil moisture content, precipitation, sea-surface wind speed, sea-surface temperature, snow cover and water content, sea ice cover, atmospheric water content, and cloud water content.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Measured signal

Surface emission

Atmospheric emission

Reflected atmospheric emission

Reflected space emission

The measured signal (TB) depends on several quantities:

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Microwave imagers

• SSM/I: -19, 22, 37, and 85 GHz- at 53°, both H and V polarizations

• AMSR: - 6, 10, 19, 22, 37, and 85 GHz- at 53°, both H and V polarizations

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

SSM/I consists of seven separate total-power radiometers, each simultaneously measuring the microwave emission coming from the Earth and the intervening atmosphere. Dual-polarization measurements are taken at 19.35, 37.0, and 85.5 GHz, and only vertical polarization is observed at 22.235 GHz. Spatial resolutions vary with frequency. The table below gives the frequencies, polarizations and temporal and spatial resolutions of the seven channels.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

absorption by oxygen and water vapour at microwave frequencies. The AMSU channels are indicated with arrows and numbers.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Typical frequencies for microwave imagers

Microwave imagers

The main aim is the investigation of surface properties

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

reflection absorption transmission

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

TS = 300º K, 37 GHz RadiometerIncidence angle = 50°

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Effects of vegetation

When vegetation is present, there are two sources of emissions – vegetation and the soil under the vegetation

Tsoil >> Tveg

Vegetation serves to attenuation the soil signature

Thus, as vegetation density increases, soil contribution decreases

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

A frontal systemobserved at 37GHzby SSM/I over theAtlantic (11/10/87-11/11/87)

Over ocean, the front isclearly observed.

Over land, lost of contrast between the land and the atmospheric contribution

© Catherine Prigent Catherine Prigent, Observatoire Observatoire de de Paris

Good reasons not to use the microwave observations over land:• emissivity often close to 1 => limited contrast with the atmospheric contribution (ocean emissivity = 0.5)• high spatial variability (vegetation, humidity, roughness, snow…)• very difficult to model on a global basisGood reasons to try:• can help characterize the land surface• if the land surface emissivity is accurately estimated, possible to estimate the atmospheric parameters over land

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

emissivity examples(SSM/I)

Vegetation:Large emissivitiesLow polarization differences

Desert:Low emissivitiesLarge polarization differences

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

December 1972 – Nimbus 5 launched with the Electrically Scanned Microwave Radiometer (ESMR) on board.- The first successful microwave imager in space.- Original motive: Mapping global rainfall rates.- Prime motive evolving after launch: Mapping global sea ice coverage.

The Nimbus Microwave Imager Heritage

103 antenna sticks with independent phase shifters Phase shifters driven with phase differences to achieve cross-track scanning at nadir, with no moving parts. Single-channel, horizontally-polarized operation at a wave-length of 1.55 cm.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

•This was the first and only time the Weddell Polynya was ever observed.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

October 1978 – Nimbus 7 launched with the Scanning Multichannel Microwave Radiometer (SMMR) on board.- Multiple channels permitted distinguishing between first-year and multiyear sea ice types.- Nearly nine-year lifetime allowed SMMR to be a major contributor to the present 32-year sea ice record.- Multiple channels also permitted observation of global sea surface temperatures independent of cloud cover.

The Nimbus Microwave Imager Heritage

There are 10 channels receiving vertically and horizontally polarized radiances at 5 different wavelengths: 0.8, 1.4, 1.7, 2.7, and 4.6 cm

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

-Greatly improves on ESMR’s ability to obtain sea ice concentrations.-Permits distinguishing between first-year and multiyear sea ice.-Obtains sea surface temperatures globally through clouds.-Determines depth of snow cover in large water basins.-Measures amount of near-surface water in the soil of large agricultural areas.-Can determine cloud water content and atmospheric water vapor.-Makes observations of global rainfall rates.

What can SMMR do with its ten channels?

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

24 years of multiyear sea ice observations

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

Special Sensor/Microwave Imager (SSMI)- A multispectral imager with similar channels to SMMR, but no 4.6 cm channel used for SST measurements- Flown on three different DOD/DMSP satellites- One month overlap with SMMR data - Has extended the sea ice record to 32 years

Special Sensor/Microwave Imager (SSMI)- A multispectral imager with similar channels to SMMR, but no 4.6 cm channel used for SST measurements- Flown on three different DOD/DMSP satellites- One-month overlap with SMMR data- Has extended the sea ice record to 32 yearsAdvanced Microwave Scanning Radiometer (AMSR-E)- Launched in 2002- Contains channels similar to SMMR plus two 3 mm channels- Has about twice the spatial resolution of SMMR & SSMI

Continuing the tradition

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

SSM/I uses several of its channels to determine brightness temperatures over land and sea. This image shows the variation of such temperatures (in degrees Kelvin) over land in the southern U.S. and the waters of the Gulf of Mexico.

Lecture on atmospheric remote sensing thomas.wagner@iup.uni-heidelberg.de

7 years precipitation climatology of TRMM/TMI processed in the GSMaP project

Tomoo Ushio et al.

top related