Satellite observations of the atmosphere and the ocean surface Heraeus Summer School

1
Satellite observations of the atmosphere and the
ocean surface Heraeus Summer School Physics of
the Environment

• Andreas Richter
• Institute of Environmental Physics
• University of Bremen
• tel. [phone]
• e-mail richter_at_iup.physik.uni-bremen.de
• http//www.iup.physik.uni-bremen.de/doas

2
Lecture Contents

• What is Remote Sensing?
• Which Quantities can be Measured?
• What are the Underlying Physical Principles?
• Examples
• Tropospheric Aerosols
• Stratospheric Ozone
• Tropospheric NO2
• Stratospheric Aerosols
• Temperature Profiles
• Wind Speed and Direction
• Sea Surface Temperature
• Sea Ice
• Summary

3
What is Remote Sensing?

• Remote sensing is the science and art of
  obtaining information about an object, area, or
  phenomenon through the analysis of data acquired
  by a device that is not in contact with the
  object, area, or phenomenon under investigation
  
  (Lillesand and Kiefer 1987)
• The art of dividing up the world into little
  multi-coloured squares and then playing computer
  games with them to release unbelievable potential
  that's always just out of reach. (Jon
  Huntington, Commonwealth Scientific and
  Industrial Research
  Organisation Exploration,
  Geoscience, Australia)

4
The Eye as a Remote Sensing Instrument

• eye remote sensing instrument in the visible
  wavelength region (350 - 750 nm)
• signal processing in the eye and in the brain
• colour (RGB) and relative intensity are used to
  identify surface types
• large data base and neuronal network used to
  derive object properties

5
The Eye as a Remote Sensing Instrument

• eyes are scanning the environment with up to 60
  frames per second
• 170 field of view, 30 focus

6
The Eye as a Remote Sensing Instrument

• stereographic view, image processing, and a large
  data base enables detection of size, distance,
  and movement

7
The Eye as a Remote Sensing Instrument

• passive remote sensing instrument, relies on
  (sun) light scattered from the object
• no sensitivity to thermal emission of objects

?
8-14 microns image of a cat
8
The Eye as a Remote Sensing Instrument

• active remote sensing by use of artificial light
  sources

?
9
Why should we use Remote Sensing?

• not all measurement locations are accessible
  (atmosphere, ice, ocean)
• remote sensing facilitates creation of long time
  series and extended measurement areas
• for many phenomena, global measurements are
  needed
• remote sensing measurements usually can be
  automated
• often, several parameters can be measured at the
  same time
• on a per measurement basis, remote sensing
  measurements usually are less expensive than
  in-situ measurements

10
Why NOT to use Remote Sensing

• remote sensing measurements are always indirect
  measurements
• the electromagnetic signal is often affected by
  more things than just the quantity to be measured
• usually, additional assumptions and models are
  needed for the interpretation of the measurements
• usually, the measurement area / volume is
  relatively large
• validation of remote sensing measurements is a
  major task and often not possible in a strict
  sense
• estimation of the errors of a remote sensing
  measurement often is difficult

11
Schematic of Remote Sensing Observation

12
Classification of Remote Sensing Techniques

• active / passive
• platform
• wavelength range
• spectral resolution
• low / medium / high
• spatial resolution
• low / high
• detection technique
• absorption, emission or extinction spectroscopy
• spectral reflectance

13
Active vs. Passive Remote Sensing

• Active Remote Sensing
• Artificial source of radiation, the reflected or
  scattered signal is analysed
• sound SONAR
• radio waves RADAR (RAdio Detection And Ranging)
• laser light LIDAR (LIght Detection And Ranging)
• white light long path DOAS (Differential Optical
  Absorption Spectroscopy)
• Passive Remote Sensing
• Natural sources of radiation, the attenuated,
  reflected, scattered, or emitted radiation is
  analysed
• solar light
• lunar light
• stellar light
• thermal emission

14
Remote Sensing Platforms

• ground-based measurements
• continuous, high accuracy, easy accessibility
• local measurement
• air-borne measurements (up to 15 km)
• fast moving, long distance
• expensive, sporadic
• sonde / balloon measurements (up to 30 km)
• high altitude
• logistically difficult, often expensive
• rocket measurements (up to 80 km)
• very high altitude
• expensive, sporadic
• Space Shuttle / Space Station measurements
• global coverage, limited time coverage, good
  accessibility
• satellite measurements
• global coverage
• poor accessibility, expensive

15
Wavelength Ranges in Remote Sensing

• UV some absorptions profile information
• aerosols
• vis surface information (vegetation)
• some absorptions
• aerosol information
• IR temperature information
• cloud information
• water / ice distinction
• many absorptions / emissions
• profile information
• MW no problems with clouds
• ice / water contrast
• surfaces
• some emissions profile information

16
Which Quantities are Measured?

• absolute intensities in dedicated wavelength
  intervals
• intensities relative to the intensity of a
  reference source
• ratios of intensities at different wavelengths
• variations of intensities
• degree of polarisation
• phase and delay of signal

17
Which Quantities can be Determined?

• Surface
• height
• albedo
• vegetation type
• surface (water) temperature
• fires
• surface roughness
• wind speed
• water turbidity / chlorophyll concentrations
• ice cover
• ice type

• Meteorology
• pressure
• temperature
• cloud cover
• cloud top height
• cloud type
• lightning frequency
• Chemical constitution of the atmosphere
• aerosol burden
• aerosol type
• trace species

18
The Electromagnetic Spectrum

• nearly all energy on Earth is supplied by the sun
  through radiation
• wavelengths from many meters (radio waves) to nm
  (X-ray)
• small wavelength high energy
• radiation interacts with atmosphere and surface
• absorption (heating, shielding)
• excitation (energy input, chemical reactions)
• re-emission (energy balance)

19
Radiative Transfer in the Atmosphere

• Contributions
• Direct Solar Ray
• Reflection on the Surface
• Reflection from Clouds
• Scattering in the Atmosphere
• Rayleigh Scattering
• Mie Scattering
• Raman Scattering
• Absorption in the Atmosphere
• Emission in the Atmosphere
• Emission from the Surface
• Emission from Clouds

20
Radiative Transfer in the Atmosphere
Atmosphere
Absorption
Scattering
Emission
Aerosol / Molecules
21
Scattering in the Atmosphere

• Depending on the ratio of the size of the
  scattering particle (r) to the wavelength (?) of
  the radiation
• Mie parameter ? 2? r / ?,
• different regimes of atmospheric scattering can
  be distinguished.

gt different wavelengths probe different parts of
the atmosphere / surface
22
What is the Optimal Instrument?

• A compromise must be found to get the optimum
  amount of information out of the limited number
  of photons available under the given boundary
  conditions

• instrument size and price
• satellite orbit
• measurement quantity
• data rate
• measurement error

spatial coverage
spatial resolution
time resolution
vertical resolution
time coverage
spectral resolution
spectral coverage
23
Satellite Orbits

• (Near) Polar Orbit
• orbits cross close to the pole
• global measurements are possible
• low earth orbit LEO (several 100 km)
• ascending and descending branch
• special case sun-synchronous orbit
• overpass over given latitude always at the same
  local time, providing similar illumination
• for sun-synchronous orbits day and night
  branches
• Geostationary Orbit
• satellite has fixed position relative to the
  Earth
• parallel measurements in a limited area from low
  to middle latitudes
• 36 000 km flight altitude, equatorial orbit

http//www2.jpl.nasa.gov/basics/bsf5-1.htm
http//www.ccrs.nrcan.gc.ca/ccrs/learn/tutorials/f
undam/chapter2/chapter2_2_e.html
24
How can Vertical Information be Derived?

• In many atmospheric application, vertical
  profiles of quantities are needed.
• Approaches
• Vertical Scanningsequential of parallel
  measurements at different altitudesgt e.g.
  SCIAMACHY limb profiles
• Pressure / Temperature dependence of signal (e.g.
  line shape)inversion of signal using a priori
  information on e.g. vertical p-profilegt e.g.
  microwave sounding
• Saturation Effects at different wavelengths
  (frequencies)using spectral regions with
  different penetration depthsgt e.g. SBUV ozone
  profile measurements
• Time Resolved measurementsusing pulsed signals
  and photon flight time informationgt e.g. LIDAR
• Combination of different types of measurements,
  instruments or modelsgt e.g. GOME tropospheric
  NO2 measurements

25
How can Vertical Information be Derived?

• In many atmospheric application, vertical
  profiles of quantities are needed.
• Approaches
• Vertical Scanningsequential of parallel
  measurements at different altitudes

Nadir observation of scattered and reflected
light, total column determination (and O3
profile), good spatial resolution, global
coverage, good SNR Limb observation of
scattered light, stratospheric and upper
atmosphere profiles, poor spatial resolution,
near global coverage, SNR decreases with
altitude Occultation direct observation of sun
or moon at horizon, stratospheric profiles, poor
spatial resolution, limited coverage (close to
terminator), high SNR but low UV
sensitivity Limb Nadir Matching combination of
nadir and limb measurements to estimate the
tropospheric column of a trace gas
http//www.sciamachy.de
26
How can Vertical Information be Derived?

• In many atmospheric application, vertical
  profiles of quantities are needed.
• Approaches
• Pressure / Temperature dependence of signal (e.g.
  line shape)

pressure broadening
inversion
http//www.ram.uni-bremen.de/index_ram.html
27
How can Vertical Information be Derived?

• In many atmospheric application, vertical
  profiles of quantities are needed.
• Approaches
• Saturation Effects at different wavelengths
  (frequencies)Example ozone profiling in the UV
  (e.g. SBUV, GOME)Ozone absorption is increasing
  by orders of magnitude over 50 nm in the UV, and
  virtually no photons reach the surface below 300
  nm. By measuring ozone at different wavelengths,
  different sub-columns are determined gt profile

28
How can the desired signal be isolated?

• In most measurements, several effects on the
  signal interfere and need to be corrected.
• Example retrieval of NO2 by UV/vis absorption
  spectroscopy of scattered sun light
• NO2 absorption
• absorption by other species (O3, O4, H2O, ...)gt
  use of measurements at many wavelengths and
  characteristic absorption spectrum for correction
• colour of the surface (e.g. ocean colour)gt use
  of measurements at many wavelengths and
  characteristic absorption spectrum for
  correction
• scattering by aerosols
• gt fit of broad band contribution
• elastic scattering by air moleculesgt fit of
  broad band contribution
• inelastic scattering by air molecules
• gt explicit correction by modelling the effect
• gt in many cases, measurements at several
  wavelengths / frequencies help

29
Validation of Remote Sensing Measurements

• Remote Sensing measurements are indirect
  measurements, and need validation!
• The perfect validation measurements should
• measure the same quantity
• integrate over the same volume
• measure at the same time
• use an independent technique
• have higher accuracy and precision than the
  measurement to be validated
• cover a large range of geophysical conditions
• have no location bias such as measurements
• only over land,
• only during clear weather or
• mostly in the Northern Hemisphere
• not be too expensive
• gt such measurements do usually not exist!

30
Problems for Validation

• Example Stratospheric NO2 measurements from
  SCIAMACHY
• Amount of data SCIAMACHY provides about 150 000
  NO2 measurements per day or more than 50 000 000
  measurements per year. To validate even a small
  part of these data necessitates a large number of
  validation measurements
• Global coverage hardly any validation
  measurements are truly global in coverage but
  usually biased over land in NH mid-latitudes
• Averaging volume even a small SCIAMACHY ground
  pixel is 30 x 60 km2 large and at high sun
  vertically integrated over the whole atmosphere.
  Sampling this volume at 3 km resolution
  horizontally and vertically (up to 20 km) would
  take many hours in an aircraft.
• Inhomogeneity in time and space many validation
  measurements do not coincide exactly in time and
  space with the remote sensing measurement.
  Horizontal variability as well as changes over
  time often are the largest uncertainty in
  validation
• Errors of validation measurements validation
  measurements often have themselves relatively
  large random and systematic errors, in particular
  if they are remote sensing measurements (example
  neglect of temperature dependence of ozone
  cross-section in Brewer measurements,
  interference by PAN and other compounds with
  in-situ NO2 measurements, pump rate problems at
  high altitudes in ozone-sonde measurements, ...)

31
Validation Example

• Example
• Validation of SCIAMACHY NO2 total columns with
  ground-based DOAS zenith-sky measurements
• Results
• validation at several stations (latitudes)
• validation of complete seasonal cycle
• comparable measurement volume
• good agreement
• Problems
• ground-based measurements AM / PM twilight,
  SCIAMACHY at 1000 LT
• zenith-sky measurements not sensitive to
  tropospheric pollution
• zenith-sky measurement is also remote sensing
  measurement, not truly independent technique

32
LIDAR Measurements of tropospheric aerosols

• Target Quantity Tropospheric aerosol
  concentrations
• Measurement Quantity Backscatter ratio at 532 nm
  and time lag
• Instrument type LIDAR
• Instrument LITE on Space Shuttle, September 1994

33
LIDAR (LIght Detection And Ranging)

• Idea
• Use of an active system that emits light pulses
  and measures the intensity of the backscattered
  light (from air molecules, aerosols, thin clouds)
  as a function of time (optical Radar)
• Instrument
• a strong laser with short pulses
• possibly several wavelengths emitted
• a large telescope to collect the weak signal
• Measurement quantity
• time lag gives altitude information
• signal intensity gives information on
  backscattering at given altitude and extinction
  along the light path
• measurements at different wavelengths provide
  information on absorbers and aerosol types
• polarisation measurements provide information on
  phase of scatterers
• gt Very good vertical resolution can be achieved!

34
Lidar In-space Technology Experiment (LITE)

• Instrument
• flashlamp-pumped NdYAG laser
• 1064 nm, 532 nm, and 355 nm
• 1-meter diameter lightweight telescope
• PMT for 355 nm and 532 nm avalanche photodiode
  (APD) for 1064 nm
• Mission Aims
• test and demonstrate lidar measurements from
  space
• collect measurements on
• clouds
• aerosols (stratospheric tropospheric)
• surface reflectance
• Operation
• on Discovery in September 1994 as part of the
  STS-64 mission
• 53 hours operation

http//www-lite.larc.nasa.gov/index.html
35
LITE Example of Aerosol Measurements
Clouds (ITCZ)
Atlas mountains
complex aerosol layer
maritime aerosol layer

• 5 minutes of LITE data over the Sahara
• low maritime aerosol layer
• high complex aerosol layer over Sahara
• Atlas Mountains separate two regimes
• clouds close to the ITCZ

http//www-lite.larc.nasa.gov/index.html
36
UV absorption measurements of stratospheric O3

• Target Quantity Stratospheric Ozone columns
• Measurement Quantity Differential absorption of
  backscattered UV radiation
• Instrument type low resolution nadir viewing UV
  spectrometer
• Instrument TOMS (Total Ozone Mapping
  Spectrometer )

37
Total Ozone Mapping Spectrometer TOMS

• Idea
• global measurements of ozone columns using
  differential measurements in the UV
• good spatial resolution through fast measurements
• additional products (SO2, aerosols) by clever
  selection of wavelengths
• continuous measurements, long time series, high
  consistency, little changes in instrumentation gt
  trends
• The TOMS programme
• Satellite Period Orbit
• Nibus 7 Oct 78 May 93 955 km
• Meteor3 Aug 91 Dec 94
• Adeos Aug 96 Jun 97 830 km
• Earth Probe (EP) Jul 96 Dec 97 500 km
• Dec 97 today 740 km
• Wavelengths
• 380.0 339.7 331.0 317.4 312.3 308.6 nm

http//jwocky.gsfc.nasa.gov/
38
TOMS Observation of the Ozone Hole

• The Ozone Hole
• forms in the Antarctic winter / spring
• formation of Polar Stratospheric Clouds PSC in
  the extremely cold vortex
• heterogeneous activation of chlorine reservoirs
  on the cold PSC surfaces
• rapid ozone destruction by ClO and BrO as the sun
  rises
• end of ozone destruction after warming when
  chlorine is transformed back to its reservoirs
  HCl and ClONO2 and vortex air mixes with ozone
  rich air

http//jwocky.gsfc.nasa.gov/
39
UV/vis absorption measurements of tropospheric NO2

• Target Quantity Tropospheric Nitrogen Dioxide
  columns
• Measurement Quantity Differential absorption of
  backscattered radiation
• Instrument type medium resolution nadir viewing
  UV/vis spectrometer
• Instrument GOME (Global Ozone Monitoring
  Experiment) on ERS-2

40
Global Ozone Monitoring Experiment (GOME)

• Idea
• simultaneous measurements from the UV to the near
  IR
• good spectral resolution (0.2 0.4 nm)
• use of DOAS to retrieve columns of several
  species (O3, NO2, OClO, BrO, HCHO, SO2, H2O)
• use of UV wavelengths to retrieve ozone profiles
• global coverage
• Launch April 1995 on ERS-2 (sun synchronous)
• GOME successor instruments
• Instrument Satellite Launch
• SCIAMACHY ENVISAT March 2002
• OMI EOS-Aura Spring 2004
• GOME-2 Metop-1 .. Metop-3 2006 2020

http//www.iup.physik.uni-bremen.de/gome/
41
GOME tropospheric NO2 excess

• NOx plays a key role in the formation of
  photochemical ozone smog
• sources of NOx are both anthropogenic (combustion
  of fossil fuels, biomass burning) and natural
  (fires, soil emissions, lightning)
• NOx emissions are changing as result of
• changes in land use
• improvements in emission control
• economic development (e.g. China)
• GOME data provided the first global maps of
  tropospheric NO2

• Data analysis
• cloud screening
• DOAS retrieval of total slant columns
• subtraction of clean Pacific sector to derive
  tropospheric slant columns
• application of tropospheric airmass factor to
  compute tropospheric vertical column

42
UV/vis Measurements of Stratospheric Aerosols

• Target Quantity stratospheric aerosol
  concentrations
• Measurement Quantity backscattered radiation
• Instrument type solar occultation viewing UV/vis
  spectrometer
• Instrument SAGE-2 (Stratospheric Aerosol and Gas
  Experiment)

43
Stratospheric Aerosol and Gas Experiment (SAGE)

• Measurement Geometry solar occultation
• Instrument grating spectrometer with
  Si-detectors
• Spectral coverage 7 wavelengths between 385
  1020 nm
• 1020, 940, 600, 525, 453, 448 und 385 nm
• Data analysis onion peeling
• Measurement targets vertical profiles of O3,
  NO2, H2O and aerosol extinction (at 385, 453,
  525 and 1020 nm)
• Measurement range stratosphere, at low
  stratospheric aerosol loading and outside the
  tropics also the upper troposphere
• The SAGE programme
• SAM II 1978
• SAGE I 1979-1981
• SAGE II 1984 - today
• SAGE III 2001 - today
• 280 1030 nm, 1-2 nm spectral
  resolution CCD detector, lunar solar
  occultation

http//www-sage3.larc.nasa.gov/
44
SAGE Stratospheric Aerosols

• Stratospheric aerosols are dominated by volcanic
  input (H2SO4).
• Large eruptions inject ash and SO2 directly into
  the stratosphere.
• Transport towards poles within one year.
• Exponential decay over many years
• 1985 Nevado del Ruiz, Columbia
• 1990 Kelut, Indonesia
• 1991 Mt. Pinatubo

http//aerosols.larc.nasa.gov/optical_depth.html
45
Radio Occultation Measurements of Temperature
Profiles

• Target Quantity temperature profiles
• Measurement Quantity excess phase of GPS signals
• Instrument type GPS occultation
• Instrument CHAMP (CHAllenging Minisatellite
  Payload)

46
CHAMP radio occultation

• Principle
• GPS receiver observes GPS satellite during
  occultation
• high accuracy time information provides excess
  phase
• this is related to the bending angle profile a
• which depends on refractive index n
• which is a function of p, T and humidity

• good vertical resolution
• large number of measurements
• good sampling
• assumptions on 2 of the three variables necessary
• problems with critical layers

http//www.copernicus.org/EGU/acp/acpd/4/7837/acpd
-4-7837_p.pdf
47
QBO Temperature Anomalies from CHAMP Radio
Occultation

• downward propagation of temperature anomalies in
  the tropical stratosphere
• QBO (Quasi Biannual Oscillation) signal
• maximum amplitude of /- 4.5 K
• impact on stratospheric ozone columns

http//www.copernicus.org/EGU/acp/acpd/4/7837/acpd
-4-7837_p.pdf
48
Microwave Measurements of Wind Speed and Direction

• Target Quantity wind speed and direction
• Measurement Quantity reflected microwave
  intensity and polarisation
• Instrument type active microwave
• Instrument Synthetic Aperture Radar (SAR).

49
How to derive wind speed from Radar signals

• Idea Bragg-like resonance of cm-size ocean
  waves with Radar signals depends monotonically on
  surface wind speed
• gt wind speed over oceans can be determined from
  scatterometer measurements if wind direction is
  known from model or other measurements

Validation
Relationship between radar backscatter and
surface wind speed for C-band (5.3 Hz),vertical
polarization at 45 off nadir angle.
http//fermi.jhuapl.edu/sar/stormwatch/user_guide
/bealguide_072_V3.pdf
50
Wind Speed from Radarsat SAR
Polar low of 05 Feb 1998 after application of
wind algorithm, embedded in NOGAPS model wind
field (arrows).
Polar low imaged by 430 km wide swath mode
of Radarsat SAR, before application of wind
algorithm, 0602 GMT 05 Feb 1998.
http//fermi.jhuapl.edu/sar/stormwatch/user_guide/
bealguide_072_V3.pdf
51
Passive Microwave Measurements of Sea Ice

• Target Quantity sea ice coverage and type
• Measurement Quantity reflected microwave
  intensity and polarisation
• Instrument type passive microwave radiometer
• Instrument AMSR-E (Advanced Microwave Scanning
  Radiometer - EOS )

• 12 channels and 6 frequencies ranging from 6.9 to
  89.0 GHz
• two polarisations

52
Sea Ice Maps from AMSR-E

• Basic principle
• strong contrast in thermal microwave emission
  between ice and open ocean
• assumption of linear relationship between
  brightness and ice cover
• parameters
• sea ice concentration,
• surface ice temperature,
• snow depth on ice
• ice type by frequency dependence of emission

http//www.seaice.de/
53
IR Measurements of Sea Surface Temperature

• Target Quantity sea surface temperature
• Measurement Quantity emitted IR radiation
• Instrument type nadir broad band IR measurements
• Instrument AVHRR (Advanced Very High Resolution
  Radiometers)

54
Reminder El Niño La Niña

• reversal of Walker circulation
• change of direction of Trade Winds
• change of ocean upwelling
• displacement of convection areas
• link to Southern Oscillation (difference of
  surface pressure between Tahiti and Darwin)

55
Sea Surface Anomaly during El Nino Event

• Sensor AVHRR
• Technique broad band IR measurements
• Quantity sea surface temperature

• Sensor TOPEX
• Technique radar altimeter
• Quantity height

56
Summary

• Remote Sensing of atmospheric and surface
  parameters from space relies on analysis of
  electromagnetic radiation emitted / scattered /
  reflected by the atmosphere and surface
• The target quantities interact with the radiation
  through absorption, emission, scattering,
  reflection or by indirectly changing the optical
  properties
• Remote Sensing measurements provide a large
  number of parameters for atmospheric physics and
  chemistry on a global scale and often over long
  time periods
• Remote Sensing measurements are indirect
  measurements and need thorough and continuous
  validation
• Spatial and temporal resolution of the
  measurements are limited and not always
  appropriate for detailed case studies
• Technological improvements and progress in data
  algorithms will further improve the usefulness of
  satellite measurements in the future
• Remote Sensing will always be only one of many
  data sources needed to understand the Earth System