WGCM summary meeting last week in Paris: CMIP5 coordinated climate change experiments proposed by WG

1
WGCM summary (meeting last week in Paris) CMIP5
coordinated climate change experiments proposed
by WGCM/AIMES

• Gerald A. Meehl (co-chair, WGCM)
• National Center for Atmospheric Research
• Boulder, Colorado

2

• Late in the IPCC AR4 process (2006), several
  issues of relevance to WGCM emerged that needed
  further attention
• Heightened interest in short term (next several
  decades) climate change information on regional
  scales, and regional weather and climate extremes
• There was no time to address new mitigation
  scenarios being formulated by WG3
• Magnitude of carbon cycle feedback was least
  quantified uncertainty Carbon cycle components
  had matured (e.g. C4MIP) such that some modeling
  groups would soon be including carbon cycle as
  part of their standard climate change models,
  some with chemistry (first generation Earth
  System Models) this strengthened WGCM
  connection to AIMES

3
WGCM and AIMES hosted a session of the Aspen
Global Change Institute in August 2006 to
formulate a new strategy for climate change
modeling and emerging ESMs that would also
connect better to WG3 (through the IAM
Consortium) and hopefully to WG2 (through
TGICA) This workshop was designed to take better
control of the model assessment process, building
on CMIP3 and C4MIP, and the scenario formulation
process through the IAM Consortium, and make the
process community-based and not IPCC-driven
(though results from a new set of coordinated
experiments would be eligible for assessment for
AR5)
4
The proposed coordinated experiments that emerged
were not dictated by IPCC, but are being
formulated by the scientific community (us) to
address relevant science questions Since these
experiments will be the major activity of the
international climate change modeling community
over the next five years, the results will be
central for assessment for the AR5 However, the
decision in April 2008 that there will be an IPCC
AR5 means that deadlines for completion of the
experiments are related to the IPCC assessment
process and schedule (publication of the WG1 AR5
in early 2013) New model versions finalized in
late 2008-2009, model runs done in 2009-2010,
data access/collection late 2010, analysis
2011-2012
5
CMIP5 Coordinated climate change experiments
(formulated by WGCM and AIMES) to be run for
assessment in IPCC AR5 Two classes of models to
address two time frames and two sets of science
questions Decadal predictability/predictio
n (to 2035) higher resolution (50 km), no carbon
cycle, some chemistry and aerosols, single
scenario (coordination contact group joint
between WGCM and WGSIP) science questions e.g.
regional climate and extremes Long term (to 2100
and beyond) intermediate resolution (200 km),
carbon cycle, specified/ simple chemistry and
aerosols, new mitigation scenarios
representative concentration pathways (RCPs)
science questions e.g. feedbacks,
processes (connect WGCM to WGNE/GCSS via cloud
processes)
(Meehl and Hibbard, 2007 Hibbard et al., 2007)
6
CMIP5 Decadal Predictability/Prediction
Experiments
7

• Decadal predictability/prediction core model
  runs
• 1.1 10 year integrations with initial dates
  towards the end of 1960, 1965, 1970, 1975, 1980,
  1985, 1990, 1995 and 2000 and 2005
• Ensemble size of 3, optionally to be increased
  to O(10)
• Ocean initial conditions should be in some way
  representative of the observed anomalies or full
  fields for the start date
• Land, sea-ice and atmosphere initial conditions
  left to the discretion of each group
• Model run time 300 years (optionally, an
  additional 700 years)
• 1.2 Extend integrations with initial dates near
  the end of 1960, 1980 and 2005 to 30 yrs.
• Each start date to use a 3 member ensemble,
  optionally to be increased to O(10)
• Ocean initial conditions represent the observed
  anomalies or full fields.
• Model run time 180 years (optionally, an
  additional 420 years)

8
CMIP5 Long-term Experiments
All simulations except those E-driven are
forced by prescribed concentrations
Carbon cycle sees 1XCO2 (1 or RCP4.5)
Coupled carbon-cycle climate models only
9

• Organizational notes
• WGCM CMIP Panel will coordinate overall aspects
  of CMIP5 experiments, data collection/archiving
  and analysis in consultation with AIMES, IAM
  Consortium, and TGICA
• Decadal predictability/prediction experiments
  overseen by joint WGCM/WGSIP contact group
  (Stouffer, Latif, Meehl, Stockdale, Boer)
• Cloud process experiments jointly overseen by
  WGCM (Bony) and WGNE/GCSS (Jakob)
• Paleo experiments overseen by PMIP (Braconnot)
• Decadal chemistry/air quality experiments
  overseen jointly by WGCM and ACC
  (Lamarque/Shindell)

10
RCPs and forcing data ready end of 2008 Modeling
groups run experiments for AR5 assessment
2009-2010 Model data archived, analyzed and
assessed 2011-2012 AR5 WG1 assessment completed
2013 Further CMIP5 model experiments (not for
AR5) run and analyzed 2011-2013 Beyond
2013 Higher resolution ESMs (25 km atmosphere,
0.1 degree ocean coupled carbon cycle,
chemistry, aerosols, dynamic vegetation) IAMs
merged with ESMs high resolution initialized
decadal predictions with 10 km AOGCMs
11
(No Transcript)
12
Decadal prediction from a multi-model ensemble
(IPCC AR4, Ch. 10 Fig. 10.8) By averaging over a
multi-model ensemble, the decadal signal is, at
minimum, 1) the forced response to increasing
GHGs (doesnt depend much on which scenario is
used) and 2) climate change commitment But if
there are modes of decadal variability that could
be predicted, the regional skill of decadal
predictions could be increased
13
Next Mitigation/adaptation New mitigation
scenarios run with earth system models will have
implicit policy actions to target future levels
of climate change But we can only mitigate part
of the problem, and we will have to adapt to the
remaining climate change The challenge use
climate models to quantify time-evolving regional
climate changes to which human societies will
have to adapt
14
Expt 1 Diagnose climate and carbon cycle
feedbacks
emissions CO2 flux CO2 concentrations
15
Expt 2 Carbon cycle response with no climate
change Climate change (AOGCM or ESM) Carbon cycle
and compatible emissions (ESM or offline carbon
cycle model)
2 - 1 effect of climate feedbacks on
compatible emissions
Emissions
16
Expt 3 Fully coupled ESM climate change
projection Use emissions from RCP scenario,
calculate fully coupled climate system response,
compare temperature change to experiment 1 to
determine size of carbon cycle feedback in terms
of climate change
17

• Permissible Emission (CO2 in air)
• Ocean/Land
  Uptake

coupled climate and carbon cycle respond to
increasing concentrations uncoupled only
carbon cycle responds (climate doesnt see
increasing concentrations)
PgC/year
SP550
SP1000
dashed Uncoupled run
solid Coupled run
Green Fossil carbon emission (reality)
Stabilization at SP550 requires a cumulative 24
reduction of permissible emissions due to
positive carbon cycle feedback (23 for
SP1000) MIROC integrated earth system model
(Kawamiya et al.)
18

• Issues to be resolved before we leave Paris
• Finalize CMIP5 coordinated experiments
• How best to quantify carbon cycle feedback? (use
  RCPs or in context of 1 runs)
• How best to formulate/prioritize the MANY
  experiments being proposed (not only in the
  WGCM/AIMES experimental design but from other
  communities)
• Linear list of prioritized experiments
• Prioritized groupings of experiments (e.g. tier
  1, tier 2, tier 3)
• A single core group of experiments, and then a
  set of non-prioritized CMIP coordinated
  experiments and let groups choose which ones to
  perform

19

• Today
• Explanation of RCP process and status (Hibbard)
• More detail on decadal (short term) experimental
  design (Murphy/Boer)
• More detail on long term experiments and metrics
  (Taylor/Stouffer/LeQuere)
• Proposals for additional coordinated experiments
  (CFMIP, chemistry, paleo, regional)
• Information from other communities regarding
  issues involving coordinated experiments (WGNE,
  WGOMD, ice sheets, impacts/WG2, IDAG)
• Recap at end of day of things to think about
  overnight

20

• Tomorrow
• Presentation of CMIP5 experimental design based
  on discussion today
• Revise if necessary
• Short presentations from modeling groups to
  update plans for climate change modeling, and
  each groups perceptions of how they view the
  coordinated experiments and what they think they
  can do given computing and other restraints

21
Decadal Prediction Experimental Design

• Part of an international climate change modeling
  activity coordinated by WCRP (WGCM/WGSIP) and
  IGBP (AIMES)
• Results from an Aspen Global Change Institute
  session on decadal prediction
• Climate Prediction to 2030 Is it possible,
  what are the scientific issues, and how would
  those predictions be used?
• June 23-28, 2008
• (co-chairs Gerald Meehl, Ron Stouffer, Lisa
  Goddard, and James Murphy)