Pascale Braconnot

1
PMIP3/CMIP5 Simulations
http//pmip3.lsce.ipsl.fr/

• Pascale Braconnot
• with input from
• Johann Jungclaus for last millenium
• Ayako Abe-Ouchi for PCMIP
• Alan Haywood for Pliocene
• and from all the contributions on PMIP wiki!

2
PMIP3 simulations

• PMIP3/CMIP5 (WGCM) simulations (past-present-futur
  e)
• Mid-Holocene, LGM, last millenium
• Same model version than the one used for climate
  projections
• Other PMIP experiments (working groups)
• Last Integlacial (130 ka)
• Mid-Pliocene (3 Myr ago)
• Transient 130-125ka (EMICS snap shots ESM)
• Deglaciation
• 8.2 kyr BP (fresh water)
• Additional simulation uncertainties in boundary
  conditions, detailed analyses of feedbacks
• Data syntheses
• Key periods, uncertainties in reconstruction
• Use of biochemical models
• Biome4, DVGM-carbon model (PC4MIP), ocean
  tracers, ice-sheet)
• PMIP database and website (LSCE)

CMIP5 long term
3
What is new since WGCM12

• PMIP3/CMIP5 boundary conditions
• PMIP interactive wiki https//pmip3.lsce.ipsl.fr/w
  iki/doku.php/
• Scientific committee and bureau
• Requirements for PMIP3 database
• New sub-projects (about 140 in total now)
• PCMIP initiative
• Next PMIP workshop in Kyoto (6-10 Dec 2010)

As soon as the boundary conditions finalized we
will contact the different modeling groups to
better know which simulations they will finally do
4
Mid-Holocene 6kyr BP
5
LGM 21 kyr BP
Vegetation map not finalized yet Ongoing
discussion about the ice-sheet
6
PMIP and ice-sheet recontructions
PMIPI
PMIPII
Peltier 2004
ICE-5G Too high in West Antarctica pb for
model-data comparisons
7
2 possibilities pb of timing

• MOCA ice-sheet from lev tarasof
• Multiparameters tuning of an ice-sheet model to
  match both geological and glaciological
  constraint.
• Use Peltiers ice sheet for geomorphology
• Provide ice extent, orography
• Provide in addition river runoff (needed for
  coupled models)
• Will be used for PMIP3 deglaciation, should be
  used for LGM
• Need at least an additional 8 weeks to be ready
  and checked
• ICE-6G from Peltier
• - Use additional constraints compared to ICE-5G
• - Should correct elevation in Antarctica
• - Ongoing discussion from the community on PMIP
  wiki
• - Should be slightly modified to take into
  account the major comments (time needed)

8
The new ICE-6G
NH
SH
21 ka 0 ka
21 ka 6G 5G)
9
Alternatives for CMIP5

• Keep ICE-5G NO
• Take ICE-6G Reasonable
• Already there
• need to see how long it could take for slight
  modifications to fit experts views
• Modeling groups will have to manage by themselves
  for the LGM river pathways.
• Take MOCA may delay the simulations
• Has the advantage to provide also river pathways
• Consistent with PMIP3 deglaciation project
• Based on all the expert comments, a decision
  should be taken by mid-October.
• All advices are welcome.

10
The last Millennium in PMIP3/CMIP5
IPCC, 2007

• Objectives
• Apply CMIP5 climate models to simulate the recent
  past
• To assess the relative role of external forcing
  and internal variability in shaping the climate
  on interdecadal to multi-centennial time scales
  ensuring conformability with the CMIP5 control
  and 20th century integrations

11
General experimental set-up
Transient expmt. 850AD -1850AD if the 1850AD
state is used as initial condition for CMIP5
20th-century runs Otherwise (e.g. lower
resolution) expmt. 850AD -2100 AD in the
Millennium set up switching to boundary
conditions compatible with CMIP5 after 1850
12
Last Millennium BC
13
Millennium boundary conditions

• Orbital parameters Table provided for 0-2100 AD
  (G. Schmidt) or internally calculated
• Trace gases Table provided 850-1850 (F. Joos)
• Volcanic aerosols Datasets available for AOD,
  effective radius, mass (C. Ammann, T. Crowley)
• Solar irradiance New TSI datasets available (N.
  Krivova, F. Steinhilber, E. Bard)
• Land cover changes Land-use conversion maps
  available from 800-2005 (J. Pongratz)
• Ice sheets, topography, coastline no changes to
  pre-ind. control

see https//pmip3.lsce.ipsl.fr/wiki/doku.php/
for updates!
14
Millennium action items

• Final decision on DEFAULT forcing set for THE
  MILLENNIUM experiment (that all groups must do)
• volcanic (Crowley/Ammann)
• solar (Krivova, Steinhilber, Bard)
• This is due pretty soon will be based on status
  of the reconstruction and its publication
• Further preparation of the data sets (e.g.
  include an 11-yr cycle for solar throughout the
  millennium)
• Ensure that many groups can participate
• e.g. provide TSI variation derived from the
  volcanic reconstructions for those models that
  cannot deal with aerosol optical depth etc.
• To be finalized within the next four weeks!

15
Millennium status
Some groups (e.g., IPSL, MPI-M) are running
Millennium experiments with the new external
forcings, reconstructions (not meeting all PMIP3
standards, or in lower resolution) MPI-M has
carried out (small) ensembles comparing one of
the new low-amplitude TSI data sets (0.1 TSI
change from Maunder Minimum to now) with an older
standard (0.25, (Ammann et al., 2007)).
16
MPI-M Millennium Simulations Northern Hemisphere
temperature
Standard solar forcing 0.1, ensemble mean and
spread
Solar forcing 0.25, ensemble mean and spread
Grey shading overlap of reconstructions (after
IPCC, 2007)

• MCA-LIA cooling more pronounced in experiments
  with 0.25 TSI variations
• Large spread in 0.1 TSI experiments during
  certain periods (in particular LIA)
• It is possible to reproduce temperature
  reconstructions with low TSI variations, but
  ensemble simulations may be necessary

17
IPSL and CNRM LM simulations
Multiproxy reconstruction
From french Project ESCARCEL Guiot et al, 2009
Regression with solar foring
18
PCMIP (PaleoCarbon Modelling Intercomparison
Project)

• (http//www.bridge.bris.ac.uk/projects/pcmip)

• Coordinated by A. Abe-Ouchi, P. Friedlingstein,
  S.P. Harrison and I.C. Prentice
• Focuses on the coupling btw. climate and carbon
  cycle on Quaternary time scale.
• More need of efforts to validate our GCMs/ESMs
• --gt Check whether the past really tell us about
  the future?
• Kick-off meeting in Jan. 2009 ( Totnes, UK)
• (1) Millenium diagnostic CO2
• (2) Millennium prognostic CO2
• (3) LGM diagnostic CO2
• This is the PMIP experiments as run by those
  modeling groups with an explicit carbon cycle,
  and is a Tier 2 experiment in CMIP5. PCMIP will
  be responsible for the carbon-cycle analyses of
  these simulations. PCMIP will also collect the
  data to validate the models.

19
Other activity in PMIP3 Pliocene
Extension of PMIP (part of PMIP phase 3)
Advisory panel Alan Haywood, Bette
Otto-Bliesner, Mark Chandler, Harry Dowsett,
Daniel Lunt, Marci Robinson, Nan Rosenbloom
PlioMIP phase 1 Experiment 1 - AGCM
Experiment 2 - AOGCM
Confirmed participation from 15 groups world-wide
http//geology.er.usgs.gov/eespteam/prism/prism_pl
iomip.html
http//groups.google.com/group/pliomip_prism
20
The mid-Pliocene and the IPCCa reawakening
The mid-Pliocene represents an accessible
example of a world that is similar in many
respects to what models estimate could be the
Earth of the late 21st Century. Taken together,
the average of the warmest times during the
mid-Pliocene presents a view of the equilibrium
state of a globally warmer world, in which
atmospheric CO2 concentrations were likely higher
than pre-industrial values. Jansen et al. 2007
chapter 6, pages 440 to 442.
21
USGS PRISM Reconstructions
22
Mid-Pliocene ResultsMean Annual Surface Air
Temperature (C)
HadAM3
GCMAM3
GCMAM3
HadAM3

• Major patterns of change reproduced

• Progressive warming towards
• high-latitudes

• But differences are evident (high lats)

23
Are these differences important - NHG
24
(No Transcript)