IPSWT_nos97

1
Capabilities of multi-angle polarization cloud
measurements from satellite POLDER results
Frédéric Parol and the POLDER ERB and
Cloudsscience team Laboratoire d Optique
Atmosphérique / CNRS
DECEMBER 2004
MAY 2002
End of OCTOBER 2005 ?
?
JUNE 2004
POLDER and PARASOL are very similar instruments
2
Outlines
? Context and rationale Cloud, radiation and
climate ? The POLDER instrument ? Interests of
multi-angle polarization measurements ? Cloud
properties ? Earth radiation balance ? POLDER
ERB,WV and Clouds operational products ?
Conclusion and perspectives
3
Context and rationale cloud, radiation and
climate
How clouds impact the Earths radiative energy
balance
-50Wm-2 in the SW 30 Wm-2 in the LW
4
Context and rationale cloud, radiation and
climate
Change in the TOA Cloud Radiative Forcing
associated with a CO2 doubling is still uncertain
positive feedback increases the warming
negative feedback reduces the warming
Effect of instantaneous CO2doubling Variation of
emitted LW 4 W.m-2
(IPCC 2001 from Le Treut and Mc Avaney, 2000)
5
Context and rationale cloud, radiation and
climate

• Small changes in clouds are potentially
  important for climate change
• Probably the greatest uncertainty in future
  projections of climate arises from clouds and
  their interactions with radiation (IPCC 2001)
• Clouds affect radiation through their height,
  their thickness and their radiative properties
• Satellite instruments allow for deriving and
  monitoring the radiation fluxes and/or the cloud
  properties (cloud cover, phase, temperature) at
  the global scale

6
POLDER concept and capabilities(POLarization and
Directionality of the Earths Reflectances)

• Main technical features
• Wide field of view lens
• 43 along track,
• 51 cross track
• ? 2400km x 1800km
• Filter wheel
• 8 spectral bands (443, 490, 565, 670, 763,
  765, 865, 910 nm) 1020nm for PARASOL
• Some are polarized
• CCD detector array
• 242 x 288 pixels
• ? Spatial resolution
• - full resolution ( 6x6 km2)
• - coarse resolution superpixel scale ( 20x20
  km2)

• POLDER instrument LOA/CNES
• Japanese platforms Adeos 1 Adeos 2
• and CNES micro satellite Myriad platform
• Missions
• POLDER 1 Nov 1997 June 1997
• POLDER 2 Jan 2003 Oct 2003
• PARASOL since Dec 2004

7
The multi-directional viewing capability of
POLDER
France 97/01/13
Up to 14 ? viewing angles per pixel for a single
satellite pass
8
Interests of the multiangle polarization
measurements

• Derivation of cloud properties from
  POLDER just two examples

9
Multiangle polarisation measurements and Cloud
Phase
Riedi et al, GRL, 2000
Variations of cloud polarized reflectance with
scattering angle exhibits very different features
depending on particle shapes (liquid droplets vs
ice crystals). The cloud phase algorithm uses 4
major differences between these angular
signatures.
10
Cloud thermodynamic phase from multiangle
polarization
11
Multiangle measurements a tool for testing
cloud (microphysical) models (1/3)
Optical thickness
reflectances
Spherical albedo
Departure from the model are coded in spherical
(or diffuse) albedo
12
Multiangle measurements a tool for testing
cloud microphysical models (2/3)
What's happened when all clouds are assumed to be
composed of spherical droplets ?
Liquid water clouds
Ice clouds
Overcast conditions over ocean
10µm-spheres
10µm-spheres
Departure from the model
Scattering angle (deg.)
Scattering angle (deg.)
Parol et al, IEEE, 1999
Doutriaux-Boucher et al, GRL, 2000
13
Multiangle measurements a tool for testing
cloud microphysical models (3/3)
The discrepancy observed for ice clouds decreases
when the Inhomogeneous Hexagonal crystal Model
(IHM) is used
14
Interests of the multiangle polarization
measurements
(B) Multiangle visible measurements TOA SW
albedo (and SW flux) retrievals
Radiance (or reflectance) measurement
Statistical ADM (observations)
Theoretical ADM (RT calculations)
Flux (or albedo)
15
1. The theoretical approach Multiangle
multispectral measurements

• Cloud optical thickness is retrieved under up to
  14 directions
• Directional parameters (optical thickness and
  cloud albedo) are calculated at 443 nm (or
  490nm), 670 nm and 865 nm.

16
Multispectral integration shortwave albedo

• ? 443 nm ? 200 550 nm
• ? 670 nm ? 550 700 nm
• ? 865 nm ? 700 4000 nm

3 Spectral albedo
SW ALBEDO
17
Juin 2003
Monthly mean shortwave albedo
Viollier et al, GRL, 2002
POLDER ALBEDO (x 100)
POLDER-CERES differences , 20S-20N mean -0,57
(-2,6)
18
2. The statistical approach ADMs based on
satellite-derived cloud properties
(a) Optical Depth lt 2.5
?180o (bwd)
?0o (fwd)
(Loeb et al, JAS,2000)
60o
30o
0o
30o
60o
(b) Optical Depth 18-40
Radiance measurement
TOA flux
?0o (fwd)
?180o (bwd)
60o
30o
60o
30o
0o
Rj(?0,?,?) is the ADM for the scene type j
0.5
0.7
0.9
1.1
1.3
R(?o,?,?)
19
POLDER ERB,WV and Clouds ProductsRapid
overview of the different levels of operational
cloud products
20
Products availability

• POLDER1 November 1996 to June 1997POLDER2
  April 2003 to October 2004
• PARASOL since Dec 2004
• Level1 calibrated georeferenced data
• Level2 daily products one file per orbit
  swath
• The level 2 (orbital swath) products contain
  about thirty non-directional parameters and ten
  directional parameters (for each of the 14
  viewing directions)
• Level3 monthly global products
• The level 3 (monthly) products contain about
  forty parameters
• Joint Atmosphere product (selected daily and
  monthly parameters)
• This product contains combined cloud and aerosol
  parameters
• Projection Lat-Lon 10802160 ( 1/6 )
• HDF file format

21
Data ordering interface from the CNES POLDER web
portal ... Web POLDER
http//polder.cnes.fr
22
Conclusion
POLDER is a very nice research instrument with
still no equivalent It provides  - Usual
standard cloud parameters  Cloud cover,
Cloud Optical thickness, Cloud pressure,
etc. - Original parameters  Cloud phase
index Information on shape and size of cloud
particles
Angular variability
23
Conclusion/Perspectives

• Possible contribution of POLDER (and the A-train)
  to the MT mission
• to establish a climatology of cloud properties
  in the Tropics
• frequency of occurrence of liquid water
  clouds/ice clouds
• discrimination between thick cirrus and
  semi-transparent cirrus
• multiangle measurements could be analyzed in
  order to investigate regional variability of ice
  cloud particles
• others
• to evaluate the existing ADMs or/and to
  construct new ADMs based on multi-satellite-derive
  d cloud properties
• To be discussed with the ERB people (M.
  Viollier, )