Applications and Limitations of Satellite Data

1
Applications and Limitations of Satellite Data

• Professor Ming-Dah Chou
• January 3, 2005
• Department of Atmospheric Sciences
• National Taiwan University

2
Why Satellite Observation?

• Other than cloud images, why do we need satellite
  data for regional weather and climate studies in
  Taiwan?

3
A short answer is

• For extended weather and climate forecasts,
  large-scale circulations and physical environment
  (e.g. SST, snow/ice cover) become very important.
  Large-scale circulations and physical environment
  can be best observed from satellite.?

4
Some Examples for Application of Satellite Data

• Model Initialization/Assimilation/Reanalysis
• Validation
• Improvements on model physics

5
ModelInitialization/ Assimilation/Reanalysis

• Initialization for weather forecast
• Assimilation
• Reanalysis (model satellite observation)
• Accurate and long-term Description
• of the earth-atmosphere system.

6
Validation of weather forecast and climate
simulations

• What parameters?

• Diagnostic
• Prognostic

• Clouds
• Radiative heat budgets
• Cloud radiative forcing
• Temperature
• Humidity
• SST
• Ice and snow cover
• Others

7
Model improvement

• Interaction between dynamical and physical
  processes (intra-seasonal and inter-annual
  variations)
• Tropical disturbances and air-sea interaction
  (momentum and heat fluxes)
• Interaction between monsoon dynamics,
  precipitation, and radiation.

8
Satellite Retrievals

• Solar Spectral Channels
• Thermal Infrared Channels
• Microwave Channels

9
Solar Spectral Channels

• Measurement of reflection at narrow channels
• Lack of vertical information

10
(No Transcript)
11
Information Derived

• Clouds
• Aerosols

• Fractional cover (visible channel)
• Article size (multiple channels)
• Cloud water amount (multiple channels)
• Cloud contamination problem especially thin
  cirrus clouds.
• Mostly over oceans.
• Large uncertainty over land especially over
  deserts
• Optical thickness spectral variation (multiple
  channels)
• Single scattering albedo (large uncertainty)
• Asymmetry factor (large uncertainty)

12
(No Transcript)
13
Information Derived (Continued)

• Ozone
• Land reflectivity
• Vegetation cover
• Ice/snow cover

• Total ozone amount (multiple channels)
• Spectral variation
• NDVI (Normalized Difference Vegetation Index)
• Reflection (albedo) difference of two channels
• Sudden albedo jump across green light
• Cloud contamination problem
• Multiple channels to differentiate clouds and ice/

14
Thermal Infrared Channels

• Rationale emission and absorption of thermal IR

15
(No Transcript)
16
Information Derived

• Temperature profile
• Water vapor profile

• Multiple channels in the CO2 absorption band
• Uniform CO2 concentration
• Weighting functions peak at different heights
• Multiple channels in the H2O absorption band
• Coupled with temperature retrievals
• Low vertical resolution
• Broad weighting function

17
Information Derived (Continued)

• Clouds

• Fractional cover
• Cloud height
• Particle size
• Cloud water amount

• Cloud-surface temperature contrast
• High spatial resolution
• Window channel
• Opaque clouds in thermal IR
• Emission at cloud top
• Unreliable
• Unreliable

18
Microwave Channels

• Emission and absorption in microwave spectrum
• Long wavelength
• Capable of penetrating through clouds

19
Information Derived

• Temperature profile
• Water vapor profile

• Multiple channels in an absorption line
• Uniform CO2 concentration
• Weighting functions peak at different heights
• Multiple channels in a H2O absorption line
• Coupled with temperature retrievals
• Low vertical resolution
• Broad weighting function

20
Information Derived (Continued)

• Precipitation

• Multiple channels
• Polarization (particle size)
• Long wavelength sensitive to large particles
• Vertical distribution of precipitation

21
SST Retrievals

• IR Technique
• Microwave Technique

22
IR Technique

• Three IR window channels (3.7, 10, and 11 µm)
• Differential water vapor absorption
• Regression
• Satellite measurements vs buoy measurements
• Sub-surface temperature
• Clear sky only
• NOAA/AVHRR, NASA/MODIS
• NOAA NCEP claims SST retrieval accuracy is
• 0.2-0.3 C

23
Microwave Technique

• Single microwave channel
• Unaffected by clouds and water vapor
• Rain (?)
• Sub-surface temperature (?)

24
Microwave Technique (Cont.)

e estimated from surface wind Ts SST Tb
Satellite measured brightness temperature
For Ts300 K and e0.5, we have Tb150K and If
?e0.001, ?Ts0.6 KVERY SENSITIVE!

• Bias among MODIS-, AVHRR-, and TRMM-derived SST
  is large, reaching 0.5-1.0 C

25
Clouds Retrieval

• Day Use both solar and thermal IR channels
• Night Use only thermal IR channels
• High spatial resolution of satellite measurements
• A field-of-view picture element (pixel) is
  either totally cloud covered or totally cloud
  free
• Cloud detection
• asat ath Tsat
• Threshold albedo (ath) and brightness
  temperature (Tth) are empirically determined

26
Clouds Retrieval (cont.)

• Zonally-averaged cloud cover of NASA/ISCCP,
  NASA/MODIS, and NOAA/NESDIS could differ by
  30-40
• Uncertainties of cloud optical thickness,
  particle size and water content are even larger
  than that of cloud cover
• Regardless of the large uncertainties of cloud
  retrievals, global cloud data sets could be
  useful depending on applications.

27
Aerosols

• Various sources/types of aerosols
• Fossil fuel combustions, dust, smoke, sea
  salt
• Large temporal and regional variations
• Short life time, 10 days
• Difficult to differentiate between aerosols and
  thin cirrus
• Difficult to retrieve aerosol properties over
  land
• high surface albedo
• Differences between various data sets of
  satellite-retrieved, as well as model-calculated
  aerosol optical thickness are large.
• Impact of aerosols on thermal IR is neglected.
• Potentially, aerosols could have a large impact
  on regional and global climate.

28
Thin Cirrus CloudsUpper Tropospheric Water Vapor

• Climatically very important
• Thin cirrus clouds are wide spread, but too thin
  to be reliably detected
• Upper tropospheric water vapor is too small to be
  reliably retrieved
• Thin cirrus clouds
• Upper tropospheric water vapor
• Although difficult to retrieve from satellite
  measurements, there are no other alternatives.
• Key to understand feedback mechanisms in climate
  change studies.

• Weak absorption visible channel (0.55 µm)
• Strong absorption near-IR channel (1.36 µm)

• Strong absorption water vapor channel (6.3 µm)

29
(No Transcript)
30
(No Transcript)