Transcript Eyring_WGCMCMIP5_091028

Report of the 13th Session of the JSC/CLIVAR
Working Group on Coupled Modelling (WGCM)
San Francisco, 28-30 September 2009
Veronika Eyring (DLR, Germany)
13th Session of the JSC/CLIVAR
Working Group on Coupled Modelling (WGCM)
San Francisco, 28-30 September 2009
OUTLINE
I. Background and goals of the WGCM meeting
II. Status CMIP5

Participating Models
 CMIP5 Simulations
III.Forcings CMIP5 simulations

Historical non-CO2 emission

IPCC Representative Concentration Pathways (RCPs)
 AC&C/SPARC Ozone Database for CMIP5
IV. Observations for CMIP5

WOAP
 NASA initiative
V. Evaluation of models

WCRP survey

Process-oriented evaluation
I. Background WGCM and Gaols of the Meeting
Background:
•
WCRP Working Group on Coupled Modelling (WGCM) leads the
development of coupled ocean/atmosphere/land models used for climate
studies on longer time-scales.
•
WGCM is also WCRP's link to the Earth system modelling in IGBP's
Analysis, Integration and Modeling of the Earth System (AIMES) and to
the Intergovernmental Panel on Climate Change (IPCC).
Members:
S. Bony and. J. Meehl (co-chairs)
P. Braconnot, V. Eyring (SPARC, AC&C), D. Karoly, A. Hirst, M. A.
Giorgetta, M. Kimoto, B. Wang, F. Giorgi N. Nakicenovic, C. Senior
Goals of this years WGCM meeting:
1. Make progress with CMIP5
2. Model evaluation
3. 1- day jointly with AIMES
II. Status CMIP5
Participating Models
Primary
Group
Country
Primary Contact
NERSC
Norway
M. Bentsen, H. Drange
Hadley Centre
U.K.
M. Collins, C. Jones
GFDL
U.S.A.
T. Delworth, I. Held, L.
Horowitz, R. Stouffer
IPSL & LMD
France
J-L. Dufresne, S. Bony
NIES & U.
Tokyo,
Japan
S. Emori, M.
Kawamiya, M. Kimoto,
CCCMA
Canada
G. Flato
MPI-HH
Germany
M. Giorgetta
INGV
Italy
S. Gualdi
EC-Earth
consortium
Europe
CSIRO &
BMRC
Australia
T. Hirst, K. Puri
NASA GSFC
U.S.A.
M. Suarez
W. Hazeleger
Primary
Group
Country
Primary Contact
CSIRO &
QCCCE
Australia
L. Rotstayn, J. Syktus, S.
Jeffrey
NCAR
U.S.A.
J. Hurrell, J. Meehl
MRI
Japan
M. Kimoto
METRI (with
Hadley Centre)
Korea
W-T. Kwon
LASG IAP
China
T. Zhou, B. Wang
NASA GISS
U.S.A.
G. Schmidt
BCC
China
Q. Li, Y. You, Z. Wang, T.
Wu, Y. Xu,
INM
Russia
E. Volodin
CERFACS &
CNRM
France
L. Terray, D. Salas-Melia
U. Reading
U.K.
L. Shaffrey
II. CMIP5 model simulations
Two classes of models to address two time
frames and two sets of science questions
1.
Near-Term (2005-2030)
high resolution (perhaps
0.5°), no carbon cycle,
some chemistry and
aerosols, single scenario
Science question: e.g.
regional extremes
1.
Longer term (to 2100 and
beyond) lower resolution
(roughly 1.5°), carbon
cycle, specified or simple
chemistry and aerosols,
benchmark stabilization
concentration scenarios;
Science question: e.g.
feedbacks.
© Crown copyright Met Office
Long-term simulations
III. Forcings CMIP5
Forcing data available on CMIP website (or via links)
http://cmip-pcmdi.llnl.gov
•
Solar (based primarily on Lean but spectrally resolved, or not)
•
Historical non-CO2 emissions
•
RCP emissions (different IAMs used to produce each RCP)
•
Land-use (U. of New Hampshire – Chini, Hurtt, Frolking)
•
Ozone time-evolving 3D historical concentrations (AC&C/SPARC)
•
AMIP SSTs and sea ice (PCMDI)
•
CFMIP Aqua-planet and idealized future pattern of SST (Hadley
Centre)
IIIa. AC&C: Historical Emissions for CMIP5
International effort to provide improved emissions 1850-2300,
consistent across 2000 for anthropogenic (including shipping and
aircraft) and biomass burning of reactive gases (not ODSs) and aerosols
Historical (1850-2000) gridded anthropogenic and biomass burning
emissions of reactive gases and aerosols: methodology and application.
Jean-François Lamarque, Claire Granier, Tami C. Bond, O. Cooper,.Veronika
Eyring, Angelika Heil, Mikiko Kainuma, Z. Klimont, David Lee, Catherine Liousse,
J. R. McConnell , Aude Mieville, S. Oltmans, Bethan Owen, D. Parrish. Keywan
Riahi, Martin Schultz, Drew Shindell, Steven Smith, Elke Stehfest, Allison
Thomson, John Van Aardenne, Detlef Van Vuuren
IIIb. Scenarios and Harmonization

Objective to provide consistent set of emissions, concentrations and
land use data at grid level for 1700-2100 (2300 period)

Harmonized in 2000 with “databases” and smooth transition to
historical trend and future scenario.

Integrated Assessment Models (IAMs) provide concentrations for
well-mixed gases, and emissions for air pollutants. Emissions will be
translated into concentrations by atmospheric chemistry.
Source: van Vuuren et al., 2009
IIIb. RCPs
RCP6.0 (not yet
available). Either keeping
constant or ramping
back to RCP4.5
Source: van Vuuren et al., 2009
IIIc. AC&C / SPARC Ozone Database for CMIP5
Effect of stratospheric ozone on climate
•
Ozone hole has led to a strengthening of the summertime surface westerlies at SH high latitudes
[Thompson and Solomon, 2002].
•
Ozone recovery is predicted to reverse that trend, with implications for the circulation of the
southern ocean [Son et al., 2008].
•
Effects of O3 depletion/recovery also in many other climate indicators showing its global impact.
•
CMIP3 models without any prescribed ozone changes (green), the past and future trends are the
same; whereas models with prescribed ozone depletion and ozone recovery are different
=> Need accurate representation of ozone recovery in climate projections.
Oct-Jan
DJF
DJF
DJF
Son et al.,
GRL, 2009
IIIc. AC&C / SPARC Ozone Database for CMIP5
Original plan: Building a new global ozone database
Original goal: create a new ozone database that would be available in time for
the CMIP5 modellers to use for AR5.
•
The NCAR database (Randel & Wu)
•
The NIWA database (Bodeker & Hassler)
•
The NOAA database (Rosenlof & Gray)
•
The GSFC database (Stolarski and Frith)
•
The Environment Canada database (Fioletov and McLinden)
However, we couldn't reach a consensus approach among the individual
database contributors. But this doesn’t mean that we have abandoned the
consensus ozone database project. The need for a consensus database
remains.
Different Tiers:
Tier 0:
Raw zonal mean monthly mean data
Tier 1:
Databases constructed using a regression model, no missing data,
tropopause to 50 km or higher
pole-to-pole coverage,
IIIc. AC&C / SPARC Ozone Data Sets for CMIP5
Goal: Provide a merged tropospheric / stratospheric ozone time series from
1850 to 2100 for use in CMIP5 simulations without interactive chemistry.
I. Cionni & V. Eyring (DLR), JF. Lamarque & B. Randel (NCAR)
A. Historical Database (1850-2009): CF netCDF monthly-mean lon, lat, pressure, time:month
1. Stratospheric data (Zonal means):
• Multiple linear regression analysis of SAGE I+II satellite observations and polar ozonesonde
measurements for the period 1979-2005 (Randel and Wu, JGR, 2007).
• Regression includes terms representing equivalent effective stratospheric chlorine (EESC)
and 11-year solar cycle variability.
• Extended backwards to 1850 based on the regression fits combined with extended proxy
times series of EESC and solar variability.
2. Tropospheric data (3D but decadal averages):
• Average from the Community Atmosphere Model (CAM) version 3.5 and the NASA-GISS
PUCCINI model.
• Both models simulate tropospheric and stratospheric chemistry with feedback to the
radiation and were driven by the recently available historical (1850-2000) emissions
succintly described in Lamarque et al., IGAC Newsletter, May 2009.
3. Combined stratospheric / tropospheric data (3D but underlying zonal mean in stratosphere):
• S and T are combined by merging the two data sets across the climatological tropopause, to
produce a smooth final data set.
FINAL VERSION RELEASED ON 22 SEP 2009 (see CMIP5 website, 16 files a 30 MB)
Total Ozone
compared to
other
observations
IIIc. AC&C / SPARC Ozone Data Sets for CMIP5
A. Historical Database (1850-2009)
see more plots at http://www.pa.op.dlr.de/CCMVal/AC&CSPARC_O3Database_CMIP5.html
Net Ozone Change 1979 to 2005 [%]
500 hPa July Ozone
1850-1859
1900-1909
1950-1959
2000-2009
Cionni et
al., in prep,
2009
IIIc. AC&C / SPARC Ozone Data Sets for CMIP5
B. Future Database (2010-2099)
•
•
•
Stratosphere: multi-model CCMVal-2 mean
Troposphere: Community Atmosphere Model (CAM) version 3.5
The data from the observational core and the model time series
are combined separately for each latitude band and pressure
level using a linear regression model.
C. Combined Ozone Timeseries (1850 to 2100)
Austin, Scinocca et al., Chapter 9, SPARC CCMVal Report, 2009
Cionni et al., in prep, 2009
Background
•
•
•
•
IVa. Observations
WOAP
WCRP Observation and Assimilation Panel (WOAP)
Karl Taylor was appointed to be WGCM’s representative on WOAP.
WOAP is a coordination Panel in WCRP
It attempts to coordinate WCRP’s interests in observation-related
activities.
• In particular, WOAP is WCRP’s preferred channel for interacting with
GCOS (Global Climate Observing System)
• WOAP helps to coordinate GCOS panels (e.g., AOPC & OOPC) (Atmos. and
Ocean Observation Panels for Climate
• WOAP has strong interest in
• Improving reanalyses
• Promoting better calibration of and especially the continuity of climate
observations
=> Make more use of the existence of WOAP within WGCM and SPARC
(SPARC Data Initiative; presentation at March 2010 workshop by
Michaela or Susann?)
IVb. Observations
NASA Initiative for CMIP5 (J. Teixeira et al.)
Objective
To provide the community of researchers that will access and analyze CMIP5
model results access to analogous sets of observational data.
 Analogous sets in terms of periods, variables, temporal/spatial frequency
 This activity will be carried out in close coordination with the corresponding
CMIP5 modeling entities and activities
 It will directly engage the observational (e.g. mission and instrument)
science teams to facilitate production of the corresponding data sets.
V. Model Evaluation
(a) WCRP Model Survey
Key deficiencies of climate simulations :
- (double) ITCZ and monsoons
- internal modes of variability of the tropical atmosphere (MJO, ISO, QBO, ENSO, etc)
- (excessively strong) equatorial cold tongue ;
- (warm) SST biases in the eastern ocean basins
- troposphere-stratosphere interactions
- regional climate change responses of precipitation and soil moisture
- cloud-climate and carbon-climate feedbacks
Key deficiencies in the models' physics :
- cloud and moist processes : atmospheric convection, precipitation, clouds in PBL, UTLS, polar..
- land-surface processes; soil moisture – precipitation interactions
- ocean-atmosphere coupling (resolution, high-wind regimes, etc)
- oceanic eddies
- non-orographic gravity-wave drag; upper boundary condition (lid)
- atmospheric chemistry
General:
- imbalance in visibility and efforts between the exploration of new, ‘hot’ territories and the
work on key persistent unresolved problems;
- the increase of models’ resolution reduces some problems, but creates new ones
- efforts put in model evaluation very unequal (e.g. climate-carbon coupled models)
- lack of inter-disciplinary interactions
V. Model Evaluation
(a) WCRP Model Survey
(1) Promote the growth of the model development community :
-> reaffirm the importance of improving basic atmospheric and oceanic components of models, ...
(2) Organize systematic and coordinated investigations (physical / statistical) of the
link between model errors and prediction errors :
- > promote systematic investigations of the impact of resolution, strato/tropo coupling, eddies..
(3) Reduce the gap between large-scale modeling/processes/observations communities :
-> encourage process-oriented evaluations/diagnostics of models (cf CFMIP, CCMVal)
(4) Reduce the gap between climate/NWP/assimilation communities
(5) Observations :
-> development of simulators for model-data comparisons
-> maintain observing network for long time series (in-situ, satellite) ...
(6) Facilitate the sharing and the distribution of ressources (cf CMIP) :
-> develop, collect and distribute diagnostics and codes (e.g. CLIVAR MJO WG)
-> facilitate access to observations and meteorological analyses
(7) Adapt the configuration of international programmes :
-> separation WCRP / IGBP : an anachronism ?
-> facilitate interactions among a large range of communities and disciplines
The results of the WCRP Survey on Model Evaluation will be written up in a e.g. BAMS paper
V. Model Evaluation
(b) Process-oriented evaluation of climate models & ESMs
IPCC, AR4
Model Intercomparison Projects (MIPs)
AMIP
AOMIP
CFMIP
CMIP
SIMIP
PILPS
C4MIP
OCMIP
CCMVal
ILAMB
MAREMIP
LUCID
AEROCOM
Is it time to get a bit more coordinated ?
Process-oriented ESM evaluation
following the CCMVal approach
Start with the evaluation of Essential Climate Variables (ECVs) from GCOS; in addition processes;
Most EU-ESM groups on board, interest from PCMDI, Article to be submitted to BAMS
Climate
Fee dback
Process
Diagnostic
Vari ables
Observations
Evaluati on
for
ES M References
OLR, profiles of T, q, xl , xi,
cloud fraction for 1xCO2
and
2xCO2
c limate
simu lations
Outgoing
long-wave
radiation (OLR)
CERES ES-4
Soden and Held [2006]
soil
–
Corre lation
between
summer evapotranspiration,
temperature
and
soil
mo isture.
Total evapotranspiration,
Sensible
heat
flu x,
Surface
te mperature,
Total and surface soil
mo isture, FAPAR
FLUXNET ecosystem sites
data, SeaWiFS & M ERIS
FAPAR
Fluxnet [2006]
Gobron et al. [1999, 2006]
Surface wind stress
forcing, formation of
intermediate
and
deep water masses
Anomalous poleward mass,
heat
and
fresh
water
transport
Hydrography
(temperature, salin ity); 3D veloc ity; volume, heat
and fresh water transports
National
oceanographic Hátún et al. [2005]
data center (NODC); inflow Østerhus et al. [2005]
and overflow transport over
sills and through openings
fro m literature
Sensitivity to changes in
CO2 and climate
GPP,
surface/leaf
temperature, prec ipitation
FLUXNET ecosystem sites
data
NDVI satellite p roduct
FACE
manipulat ive
e xperiments
Sensitivity to changes in
climate, ocean circulat ion,
and rising CO2
POC primary production,
POC e xport production,
PIC production,
POC part icle flu xes,
Ca CO3 part icle flu xes
Primary
productivity Behrenfeld and Falk owsk i
derived
fro m
re motely [1997a,b] Behrenfeld et al.
sensed
ocean colour [2005]
(SeaWiFS)
Physical climate fee dbacks
Atmospheric Dynamics & Cl ouds
Water
vapour/lapse
rate feedback
Positive
climate
feedback
by
increased
water
vapour greenhouse
effect
Land Surface physics
Land-cover
status - energy
balance
feedback
Strength
of
mo isture
temperature
coupling
Ocean Dynamics & Sea Ice
North Atlantic
thermohaline
circulat ion and
climate
feedbacks
Gl obal car bon c ycle fee dbacks
Land Biogeoc he mistry
Feedback
between
climate change
and
Net
Ecosystem
Productivity
Gross
Prima ry
Productivity
Fluxnet [2006]
Tuck er et al. [2005]
Norby et al. [2005]
Marine Biogeoche mistry
Feedbacks
between
climate change
plus rising CO2
and
the
biological
carbon pumps
Biologica l partic le
production at the sea
surface and vertical
particle
flu xes
throughout the water
column
Partic le flu xes from JGOFS
data base
Atmospheric composition fee dbacks
Aerosols
Aerosol
–
climate
interactions
and feedbacks
Oxidation,
wet
re moval & vertica l
mixing
in
troposphere
Changes in precipitation
rate, wet deposition rates,
aerosol residence
time,
vertical part itioning
Speciated wet re moval EM EP, IM PROVE, NADP,
rates,
precipitation, aircra ft and lidar profiles
mixing ratios,
(CA LIOP)
Temperature - and climatedependent changes in NOx
and VOC e mission rates
Mixing ratios of NOx and
key VOCs (especially
isoprene and products)
Rae et al. [2007]
Kirk evag et al. [2008]
Che mistry-Cli mate
Changes in RF
due
to
tropospheric
ozone
and
other oxidants
Biogenic precursor
emissions
(VOC
and NOx)
Land and oce an e missions & de position
Soil
– Soil e missions; dry Climate-dependent changes
deposition
in wind blown dust
atmospheric
chemistry
feedbacks
SCIAMACHY,
GOM E2 Yienger and Levy [1995]
(CH2 O as an indicator of Jaegle et al. [2005]
VOCs);
CM DL
flask Guenther et al. [2006]
network; composites of
fie ld ca mpaign data
Mass
and
size Satellite
dust
aerosol Tegen et al. [2004]
distributions
of
dust, products, AERONET
Balk ansk i et al. [2004]
surface wind speed, and
soil characterization, soil
mo isture and vegetation
cover
Concept of
processoriented ESM
evaluation
Proposal for an ESM MIP
E S M
I
P
Aim : Facilitate/encourage/enforce process oriented evaluation of
ESMs
- Coordinated diagnostic effort
- Build on previous MIPs experience
- Focus on processes and feedbacks relevant for climate projections
- Use of global 20th century observations
- Use of CMIP5 and related model simulations
- ESM perspective (eg. coupling issues)
- To be endorsed by WGCM and AIMES ?