HOW TO USE ICE CLOUD PROPERTIES DERIVED FROM THE ATRAIN TO HELP MODELLERS

1
HOW TO USEICE CLOUD PROPERTIES DERIVED FROM THE
A-TRAIN TO HELP MODELLERS
Cloud and Radiation Collaboration Meeting- Exeter
2008
Improve climate forecasts
GCMs

• Julien Delanoë Robin Hogan
• University of Reading, UK
• Thanks to
• Richard Forbes (ECMWF)
• Alejandro Bodas-Salcedo (Met-Office)

Cloud and Radiation Collaboration Meeting
September 17-18th 2008
2
Introduction
Cloud and Radiation Collaboration Meeting- Exeter
2008

• The main (tricky) question how can observations
  help modellers to improve their models?
• We can provide cloud properties all around the
  globe (with as much or as little detailed as
  required)
• Compare observation to model output (IWC/re,
  IWP/optical thickness)
• Try to improve cloud parameterization,
  statistical relationships as a function of key
  model parameters, cloud types etc
• Issues? Things that we have to be aware about
• Instrument sensitivity and limits (cant see
  everything)
• Algorithm assumptions
• Simulate instruments from model ouputs and
  compare
• Wilkinson et al. (lidar)
• Bodas-Salcedo et al. (radar)

3
Global data
Cloud and Radiation Collaboration Meeting- Exeter
2008
A-Train 28th April 2006

• CloudSat Cloud profiler radar 94GHz
• CALIPSO Cloud profiler lidar 532, 1064nm Infra
  Red Imager
• AQUA radiometers MODIS, AIRS, CERES, AMSR-E

Global coverage Radar 2.5 km along track X 1.2
km across track / 500m gt250m Lidar 333 m / 30
m Our merged product CloudSat footprint/vertical
resolution 60m
4
Cloud and Radiation Collaboration Meeting- Exeter
2008
Why combine radar, lidar and radiometers?

• Radar Z?D6, lidar b?D2 so the combination
  provides particle size
• Lidar sensitive to particle concentration, can
  be extinguished
• Radar very sensitive to the particle size, not
  very sensitivity to liquid clouds and small ice
  particles

CALIPSO lidar
CloudSat radar

• Radiances ensure that the retrieved profiles can
  be used for radiative transfer studies
• Single channel information on extinction near
  cloud top
• Pair of channels ice particle size information
  near cloud top
• We use unified variational scheme to retrieve
  ice cloud properties, thin and thick ice clouds
• Delanoë and Hogan 2008, JGR (doi10.1029/2007JD009
  000)

5
Formulation of the scheme
Cloud and Radiation Collaboration Meeting- Exeter
2008
We know the observations (instrument
measurements) and we would like to know cloud
properties visible extinction, Ice water
content, effective radius

• Observation vector State vector (which
  we want to retrieve)
• Elements may be missing

Iterative process compare predicted observations
and measurements, with an a-priori and
measurement errors as a constraint
6
Important assumptions
Cloud and Radiation Collaboration Meeting- Exeter
2008

• Assumptions
• Density-Area-diameter relationship Brown and
  Francis (1995)
• N0/extinctioncoeff temperature dependency (from
  aircraft data, 2DC-2DP etc)
• Normalization of N(D) by N0
• Lookup tables to convert
• extinction/N0 to IWC and re

CALIPSO lidar
Only radar IWC retrieval is similar to IWC-Z-T
relationship (Hogan et al. 2005,Protat et al.
2006) Radar/lidar good confidence in the
retrieval Only lidar IWC can be poor retrieved
due to the small particle issue. However
extinction very good.
CloudSat radar
Ice water content
No obvious discontinuities between different
parts (radar/lidar/radar-lidar)
7
Cloud and Radiation Collaboration Meeting- Exeter
2008
Ice cloud properties from A-TRAINCloudSat-CALIPS
OStatistics one month data July 2006Ice water
content and optical thickness
8
Cloud and Radiation Collaboration Meeting- Exeter
2008
Frequency of occurrence of IWC vs temperature

• IWC increases with temperature
• but spread over 2 to 3 orders of magnitude at
  low temperatures
• reach 5 orders of magnitude close to 0 C

• Advantage of the algorithm
• Deep ice clouds radar
• Thin ice clouds lidar
• When radar and lidar work well together very good
  confidence in the retrievals
• Obvious complementarity
• radar-lidar

9
Cloud and Radiation Collaboration Meeting- Exeter
2008
Differences between Hemispheres July 2006
No obvious differences in the general trend IWC
shifted to low temperatures in southern
hemisphere
Temperature C
Temperature C
Boreal Summer
Austral Winter
10
Comparison of ice water path
Cloud and Radiation Collaboration Meeting- Exeter
2008

• Mean of all skies
• Mean of clouds

CloudSat-CALIPSO
MODIS
Need longer period than just one month (July
2006) to obtain adequate statistics from poorer
sampling of radar and lidar
11
Cloud and Radiation Collaboration Meeting- Exeter
2008
Ice cloud properties from A-TRAINCloudSat-CALIPS
OModel comparisons (July 2006)UK
Met-OfficeECMWF
12
Cloud and Radiation Collaboration Meeting- Exeter
2008
Forecasts Vertical profiles were extracted from
the model along the CloudSat-Calipso tracks at
the closest time to the observations A-train
data averaged to models grid
Good trend but cant capture variability
lack of IWC _at_ Tlt -50C
gt Models capture the trend of the IWC-T
distribution (not the rest)
13
Cloud and Radiation Collaboration Meeting- Exeter
2008
Ice cloud properties from A-TRAINCloudSat-CALIPS
OModel comparison (July 2006)UK
Met-OfficeDoes the model capture the latitude
characteristics?
14
Cloud and Radiation Collaboration Meeting- Exeter
2008
A-Train vs UK met-Office
Northern Hemisphere
Southern Hemisphere
A-Train
A-Train
UK met-Office
UK met-Office
Temperature C
log10(IWC)
log10(IWC)
log10(IWC)
log10(IWC)
Ok for the trend but not the variability
15
What do we do?
Cloud and Radiation Collaboration Meeting- Exeter
2008

• Right now we can provide (1 month July)
• Ice cloud property profiles combining radar,
  lidar, radiometer
• Develop ice cloud climatology and use to evaluate
  forecast/climate models
• Very soon should have few months (ICARE will
  provide radar-lidar merged files for all the
  CloudSat-CALIPSO periodhttp//www-icare.univ-lill
  e1.fr/)
• What we could do
• Statistics as function of
• Type of clouds (stratiform/cirriform/convective)
• Locations (any priority, critical areas)
• What do you think, what would help (interest)
  you?
• What is the best way to present these data to
  diagnose specific errors?
• What kind of relationships?
• What are currently the weaknesses in GCM?
• Acknowledgements These data were obtained from
  the NASA Langley Research Center Atmospheric
  Science Data Center and the NASA CloudSat
  project.

16
backup
17
Comparison of optical depth
Cloud and Radiation Collaboration Meeting- Exeter
2008

• Mean of all skies
• Mean of clouds

CloudSat-CALIPSO
MODIS
Mean optical depth from CloudSat-CALIPSO is lower
than MODIS simply because CALIPSO detected many
more optically thin clouds not seen by MODIS gt
need to compare PDFs as well
18
Attenuated lidar backscatter from CALIPSO
Radar Reflectivity from CloudSat
We have developed a simple cloud phase
identification algorithm
Cloud seen by lidar
Temperature model (ECMWF) gt Ice / Liquid
water Simple method Exploit the different
response of radar and lidar in presence of
supercooled liquid water -Very strong lidar
signal -Very weak radar signal Within a 300m
cloud layer
Cloud seen by radar
19
Northern hemisphere
July 2006
Southern hemisphere