Transcript Cloud Feedbacks CPT

Low-Latitude Cloud Feedbacks CPT
Chris Bretherton
University of Washington
• US CLIVAR activity sponsored by NSF and NOAA
at ~$1M/yr, along with 2 ocean ‘climate process
teams.’
• 1 Oct. 2003-30 Sept. 2006.
• A pilot project – can a multi-institution, multi-model
effort integrating observationalists, diagnosticians,
process-scale and large-scale modelers improve
climate models faster?
• NCAR, GFDL, GMAO(NASA) models.
Low-latitude cloud feedback CPT
Goal: Reduce uncertainty in
low-latitude cloud feedbacks
on climate sensitivity.
• In-depth diagnosis of cloud
feedbacks in models.
• Implement ‘best-practices’
parameterizations honed via
single-column methods.
• Start with boundary-layer
clouds, move to deep
convective systems.
The time is ripe…
• Cloud feedbacks on climate sensitivity are a longstanding problem.
• ~1 yr ago, CAM2 climate sensitivity was 1.5 K,
AM2.10
was ~5 K for 2xCO2
Subtropical PBL cloud feedbacks responsible for much
of this difference.
Now,
CAM3 climate sensitivity is ~2.9 K
AM2.12
is ~2.6 K
Both models have evolved, but this demands clear
explanation...can we prove newer is better?
• Over short run, diagnostic findings may help IPCC AR4.
Over longer run, IPCC model finalization allows more
thoughtful parameterization development.
CPT organization
• Core group (C. Bretherton, M. Khairoutdinov, C. Lappen,
B. Mapes, J. Norris, R. Pincus, B. Stevens, K. Xu, M.
Zhang): Parameterization, diagnosis, observational
hooks.
• Advisory group (B. Albrecht, A. Betts, C. Fairall, T. del
Genio, S. Ghan, G. McFarquhar, R. Mechoso, H. Pan, D.
Randall, D. Raymond, J. Teixeira, R. Weller)
• NCAR–Kiehl, Rasch, Collins; Liaison: hiring underway
GFDL-Klein (thru 3/04), Held, Donner; Liaison: Zhao
GMAO-Bacmeister, Suarez.
• Not an exclusive effort - active coordination with CCSM
AMWG (one reason for this talk!), GCSS, European,
Canadian efforts, CFMIP.
Strategy
• Hypothesis-driven diagnoses of biases, cloud
feedbacks
• Single-column methods and intercomparison .
• PI-specific efforts
- New PBL, ShCu, microphysical parameterizations
- Superparameterization cloud response to climate
change (e.g. specified dSST)
- Use of long-term cloud datasets and new satellite
obs.
- CRM/LES simulations of cloud feedbacks to specified
greenhouse-type large-scale forcing changes.
- Novel model diagnostics.
CPT Meeting 20-21 Nov, 2003 at NCAR
1. Focus teams for process diagnosis and new physics:
Cloud topped boundary layers (Bretherton)
Deep tropical convective systems (Mapes)
Representing subgrid heterogeneity in microphysics
and radiation (Pincus)
2. Multimodel simulation archive
3. Column output locations
4. Group projects
Rad.-conv. equilibrium intercomparison
Intercomp. of CTBL response to different climate
forcing components (e.g. changed free-tropospheric
subsidence and humidity, LTS)
Cloud response mechanisms to climate change
• Deep convective clouds
- Freezing level increase
- Anvil top temperature feedbacks (Hartmann)
- LWCF vs. SWCF – is cancellation universal?
• PBL clouds:
- Stronger trade inversion due to stabler moist adiabat
- More emissive free troposphere.
- Subsidence changes.
Simulation archive
Monthly climatology history files from recent
versions of models suitable for climate
sensitivity analysis.
GFDL
NCAR
GMAO
AM2.12 (59e) CAM3(rio33) NSIPP
ctrl
y
y
y
dSST +2/[-2]
y
y
y
dSST_CMIP
y
y
y
SOM (1x,2x CO2)
y
y
Single column model y
y
Some sample diagnostics so far…
• Climatologies
• Bony-grams (binning vs. 500 mb w)
ERA15
Control
runs
Annual mean W(500 hPa)
w(500 mb)
(mb d-1)
NCAR
Annual mean
Control run
GFDL
GMAO
ERBE
-SWCF
(W m-2)
NCAR
Annual mean
Control run
GFDL
GMAO
NCAR
SST + 2
Bonygram
NCAR
GFDL
GMAO
SST+2
Bonygram
comparison
NCAR
SST+2
vs.
2xCO2 SOM
NCAR
SST+2
vs.
SOM 2xCO2
Summary
• CPT is spinning up.
• Regular reports at AMWG meetings
• Many opportunities for collaboration with other
AMWG members.