CESM Workshop1 July 2010Jim Hurrell
Summary and Discussion12:00-12:30
• Atmosphere Model• Ocean Model• Land Model• Polar Climate• Q&A
12:30-13:00 • Land Ice• Biogeochemistry• Chemistry-Climate• WACCM• Q&A
13:00-13:30 • Software Engineering• Climate Change• Climate Variability• Paleoclimate• Q&A
13:30 Adjourn
Additional Feedback
Jim Hurrell
Atmosphere Model Working Group: SummaryCAM4Convection/polar cloud changesMean climate improves on CAM3More realistic variability
Diurnal, intraseasonal, interannualMore high-resolution experienceMore coupled experience (inc. chem)USE: Workhorse for climate applications
CAM5Complete upgrade to physics from CAM3More consistent physical systemIncludes cloud/aerosol indirect affects2-deg climate competes with CAM4 1-degHigher climate sensitivity (4K/2xCO2)More expensive (~x4)USE: Atmosphere process applications; model development
More detailed guidance on CESM1-CAM5 release web page (to appear)
Obs
CAM4
CAM3
CCSM4SST’
PBL and shallow cumulus
2-moment microphysics
Atmosphere Model Working Group: Plans
Shorter Term (aim: Use CAM5 for everything!) Understand CAM5 climate and physical process interactions further Reduce remaining biases: (a) long-wave radiation, in-cloud water
(b) strong cloud-aerosol indirect affects (20th Cen.)(c) excessive El Nino SST anomalies
Reduce cost: (a) Software engineering; (b) aerosol simplifications (prescribed burdens) An updated release of CAM5 is highly likely in the next few months to a year Observed datasets to validate and evaluate new processes (aerosols, clouds: simulators)
Longer Term Guidance provided via CAM Strategic Plan (available on the web) Towards regional climate simulation (w/ SEWG, DOE for HOMME, MPAS)
Global high resolution Regional refinementNested domains
Scale invariant/scale aware parameterizations (e.g., sub-columns) Reducing uncertainty and persistent model errors (e.g., twin ITCZ -> DOE-CSSEF) Build and maintain capacity and capability for the community (MMF, testbeds, CAPT, DART)
Ocean Model Working Group
• Results: CCSM4 control and 20th Century– Plus results from soon-to-be-released low
resolution version of coupled model
• Decadal prediction– POP has been implemented within DART
• Jeff Anderson’s team in CISL
OMWG: Strongly eddying ocean-only simulations
• Ensemble of dye release simulations from Deepwater Horizon
• Greenland fresh water perturbation experiments
OMWG: “Next Cycle” Model Development
• New Climate Process Teams– Internal waves and mixing– Ocean mixing and sea ice heterogeneity
• Immersed Boundary Method into POP– Primary purpose: Dynamic ice shelf/ocean
interface
OMWG: Longer Term Development
• DOE Plans– likely establishment of position at NCAR, to work
with LANL on
• The Model for Prediction Across Scales (MPAS)– Enstrophy dissipating (PV conserving) dynamics
• Extension of Arakawa and Lamb to highly variable grids• Voronoi tessellations (hex’s) or quadrilaterals
Advances for CLM4
– Terrestrial carbon and nitrogen cycle (CN); merge with DGVM (CNDV)
– Transient land cover/land use change including wood harvest
– Urban model
– Revised hydrology scheme
– Revised snow model (SNICAR aerosol deposition, snow cover fraction)
– Improved permafrost representation
– New surface datasets (albedo improved)
– MEGAN VOC emissions model
– …
CLM near-term activities, CLM4.x (~ 1-2 years)
– Crops and irrigation (by end of summer) – still CLM4.0
– Revised cold region hydrology, prognostic wetland distribution
– Gross Primary Productivity (update photosynthesis model)
– Improved fire algorithm including human triggers and suppression
– Revised lake model
– Methane emissions model
– High resolution River Transport Model
– Introduce dynamic landunits
• Transitions: glacier to vegetated,
primary vegetation to crop, lake area change
– Offline meteorology forcing dataset
• 1900-2009, updatable
– Participate in Global Carbon Project assessments
– Improved metrics, diagnostics
CLM medium-term development activities
– 3-D canopy radiation
– Ecosystem demography, temporal response to disturbance
– Soil carbon
– Sub-grid soil moisture and snow heterogeneity
– Isotopes (?)
– N2O emissions
– Riverine transport of nutrients and sediments
Polar Climate Working GroupCo-chairs: Marika Holland (NCAR) and Elizabeth Hunke (LANL)
NCAR is sponsored by the National Science Foundation
Model analysis of high latitude climate
• Ongoing analysis to understand and document the CCSM4/CESM1 models– Influence of new solar radiation capabilities in CICE– Low resolution CCSM4– Arctic climate feedbacks in CAM4/CAM5– Polar climate simulation and its 20th-21st century
change– Arctic ocean freshwater budgets
• In near term, this analysis be will extended and documented in J. Climate special issue papers
Sea Ice Model Developments
• CICE parameter study• Modeling ice bergs and their interaction with
sea ice• Prognostic salinity for CICE• Melt pond modeling• Brief discussions of other developments:
dynamics, biogeochemistry, snow model improvements, ice-ocean coupling
Summary of Land Ice Working Group Meeting
• 12 talks, ~55 attendees, and 4 major themes• Overview and perspective:
– IPCC meeting in Malaysia – seaRISE experiments and status
• Use of model for:– Improved surface mass balance– Basal water distribution and traction– Model initialization
• Improvements to core model:– Discretization, field equations, and meshing schemes– Solver and scalability– Sensitivity
• Data support:– Improved data bed topography– Data assimilation methods
Summary of LIWG• Achievements:
– Coupling to CESM– A group of users is addressing scientific issues including surface
mass balance, dynamic basal traction, and intialization– Developmental support from DOE ICICLES project is quickly
advancing (scalable, robust, adaptive, and better physics).• Provocative questions (raised by David Holland):
– Current efforts to make SLR predictions in time for AR5 will appear to be rushed. Since these will be the basis for policy decisions into the next decade, do we want to risk making that prediction?
– Should we push for an IPCC focus on SLR between AR5 and AR6?
BGCWG 2010 Breckenridge Summary
• Summary of CCSM4/CESM1 release– New features & out of the box compset
• Overview of 1850 controls & 20C transients– C cycle is stable in 1850 controls– Too much C stays in ATM in 20C transient
– Ocean surface BGC improved vs. CCSM3– Oxygen Minimum Zones are far too large
– CLM-CN GPP seasonal cycle improved vs. CCSM3– GPP drops 20% from CLM standalone to CESM1
BGCWG 2010 Breckenridge Summary
• Analysis– Impact of climate change on marine productivity– Importance of seasonal cycle in forcing
• Development– Impacts of prognostic aerosol/dust on climate/BGC– Spinning up CLM-CN– Soil Carbon Modeling– Methane Modeling
Chemistry-Climate Working group meeting
Arlene Fiore (GFDL)Peter Hess (Cornell)
Jean-François Lamarque (NCAR)
NCAR is sponsored by the National Science Foundation
Status of CESM1(CHEM)• Documented and released• 1850 Control: 30 years available on ESG• Transient 1850-2005 simulation completed:
planning two additional simulationsBlue: SuperfastGreen: CCSM4Black: Obs.
Next steps (1-year horizon)• Additional ensemble members of CMIP5 20th century
CESM1(CHEM) simulation• Model Development
– Improve wet deposition and photolysis rate parameterizations
• Model Evaluation– Expand diagnostic package and integrate in AMWG’s– Chemistry in new model configurations: HOMME, CAM5– Participation in International Intercomparisons:
ACC-MIP, AeroCom, Transcom-CH4, …• CPT?• New strategic plan
NCAR is sponsored by the National Science Foundation
WACCM Working Group Meeting Summary
Dan Marsh, NCAR, Boulder
Planned coupled simulations 2010
1850 control
20thC
1850
1960
2005
2050
IPCC-1Obs.
IPCC-2RCP4.5
simulation years
mod
el ti
me
1000
2050
WorldAvoided
1985
≈
200
2100
Future Plans
• Complete CMIP5 simulations
• Document model - high vs. low top & solar cycle / ozone hole effects on troposphere
• Migrate WACCM4 to CAM5 physics (esp. RRTMG)
• Integrate CARMA bin-microphysics option (UTLS and strat. aerosol studies)
• “World Avoided” simulations
• Release of WACCM-X and Specified Dynamics Versions
• Engage CEDAR community
Climate Change Working Group: 2010-2013
Co-chairs: Gerald A. Meehl, Warren M. Washington, Karl Taylor
The challenge for the CCWG: run and analyze a profusion of new models, new model configurations, model resolutions, model experiments
New model versions that will have future climate change simulations
CCSM4, 1 degree atmosphere, full suite of CMIP5 simulations
CCSM4, 2 degree atmosphere, 20th century and single member RCP simulations
CCSM4, 0.5 degree atmosphere, 100 year control, single member 20th
century and RCP4.5; decadal prediction experiments
CESM1 (WACCM) (2 degree atmosphere, stratospheric dynamics, ozone chemistry): 20th century and at least one RCP4.5 to 2100
CESM1 (CAM5); (1 degree atmosphere) pre-industrial control, 20th century and several RCP 21st century simulations
CESM1 (BGC): CMIP5 carbon cycle feedback experiments, 20th and 21st
century RCP4.5
HOMME 0.25°, T341 CCSM4, 0.25° CCSM4: atmosphere-only time slice experiments (20 yrs end of 20th century; 20 yrs end of 21st century)
CESM1 (Glimmer-CISM), 20th century, RCP4.5, RCP8.5
CMIP5 Decadal prediction experiments
Multi-ensemble member sets of CMIP5 decadal hindcast and prediction experiments with two initialization schemes:
1. Ocean-Ice Hindcast (CORE)
2. DART weakly coupled scheme (ocean data assimilation weakly coupled to atmosphere data assimilation)
Some results so far
The context:
Equilibrium climate sensitivityPCM: 2.1°CCCSM3: 2.7°CCCSM4: 3.1°CCESM1 CAM5: 4.5°C
Transient climate response (TCR)PCM: 1.30°CCCSM3: 1.46°CCCSM4: 1.56°CCESM1 CAM5: unknown at this time
CCWG Model Runs2010
CMIP5 runs
CMIP5 experiments; climate variability/change analyses; CCSM4 aerosol sensitivity experimentsCCSM4 20th-century & RCP scenariosCMIP5 Long-term stabilization scenarios; CCSM4 single forcing experimentsCESM CMIP5 20th century and RCPs
CCSM4 decadal prediction runs½° CCSM4 simulationsHigh-resolution time slices (0.25°HOMME, T341, CCSM4)
CAM-HOMME runs, CCSM4
Analysis of climate variability andclimate change; geoengineering runs; continue CMIP5 experiments;
cloud-resolving CESMHigh-resolution decadal prediction experiments; CCSM4/CESM perturbed physics ensemblesCESM single forcing experiments
CESM decadal prediction experiments;High-resolution coupled experiments;Special DOE scenarios for
future US energy strategies
CCSM4/CESM, CAM-HOMME
2011-2013CESM1 & high resolution
The Geoengineering Model Intercomparison Project, GeoMIP, a CMIP Coordinated Experiment (Ben Kravitz, Alan Robock, Olivier Boucher, Hauke Schmidt, Karl Taylor, Georgiy Stenchikov, Michael Schulz)
Aerosol radiative forcingPCM: sulfate direct ~ -0.8 Wm-2
CCSM3: sulfate direct ~ -0.8 Wm-2
CCSM4: sulfate direct -0.8 Wm-2
CESM CAM5: sulfate direct -0.5 Wm-2; indirect -1.6 Wm-2
CESM WACCM
-20
0
20
40
60
80
100
120
2000
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
Emis
sion
s (G
tCO
2)
MiniCAM 4.5
IMAGE 2.6
AIM 6.0
MES-A2R 8.5
IMAGE 2.9
300
400
500
600
700
800
900
1000
2000
2010
2020
2030
2040
2050
2060
2070
2080
2090
2100
CO2
Conc
entra
tions
(ppm
)
MiniCAM 4.5
AIM 6.0
MES-A2R 8.5
IMAGE 2.6IMAGE 2.9
Baseline range (10-90th percentile)
Stabilization range (10-90th percentile)
Post-SRES (min/max)
Representative Concentration Pathways (RCPs)
RCP8.5
RCP6
RCP4.5
RCP2.6
To 2300
RCP8.5
RCP6
RCP4.5
RCP2.6
CCWG CMIP5 experiments for assessment in IPCC AR5 Estimated Timeline
9/13
CMIP5 experiments performed
CESM Climate Variability Working Group Session
• Extremely Well Attended – Packed Room• Eight 20-Minute Presentations
– Various Versions of CCSM
• Planning for CVWG Numerical Experiments to be Conducted via email and conference calls
• Branstator– Decadal Predictability Limits in CCSM3+CCSM4
• Identical Twins, Analogues, Entropy– Limits in North Pacific and North Atlantic on order
of 5-8 years– Limits Longer in CCSM3 compared to CCSM4
• Bitz– Seasonal Prediction of Sea-Ice and Surface
Temperature in the Arctic.– There is substantial autocorrelation on seasonal
time-scales due to different mechanisms• Alexander
– Tropical-Extratropical Interactions and ENSO– North Pacific Variability as an trigger for ENSO
• Cole– Paleo-records of trends and variability in tropical
Pacific– Rich spectrum of ENSO variability from proxy data– Decadal signal relatively small in modern instrument
record (last 50 years)– Proxy records give warming trend in tropical Pacific in
recent years• Clement
– Rethinking the role of the Ocean in the Southern Oscillation
– Southern Oscillation red spectrum independent of ocean dynamics (Atmos+SOM)
– Interannual peak due to Bjerknes feedback and ocean dynamics
– Decadal variability amplitude reduced by ocean dynamics
• Lu– The Role of the Ocean Dynamical Feedback in the Climate
Response to GHG Warming– Flux override experimental design presented– Wind stress in 2xCO2 leads to tropical Pacific cooling
• Schneider– How weather noise impacts the forced climate response– Interactive ensemble coupling strategy– IE captures global mean forced response– Large regional differences in forced response associated
with noise reduction• Anderson
– On the Radiative Quasi-equilibrium between SST and Anthropogenic Climate Forcing
– AMIP style experimental protocol to isolate known radiative forced response and implied ocean response
– Ocean natural variability and unknown radiative forcing unlikely to explain warming trend of the last 50 years
Paleoclimate Working Group
CESM - Challenges for an Earth System Model
How can we utilize information that integrates across compartments and disciplines
• to better enjoy and understand the interacting nature of the Earth System
• to remain competitive in the international arena
• to profit from new programmes, like GEOTRACERS
Increase efforts to incorporate in the core of CESMadditional oxygen, water and carbon tracers
• that quantify the cycling of water through the ocean –atmosphere – cloud - sea ice - land ice -soil system
• that reveal quantitative insights into the source mix of carbon dioxide, methane and other GHGs
• that provide metrics on the transport time scales within and between the ocean, atmosphere and land
• that reveal insights into the interplay between atmospheric chemistry and transport
• that improve the interpretation of the paleo proxy records for improved predictability of future change
• ....
CESM - Challenges for an Earth System ModelContributions from the Paleoclimate WG
• Contribute to the incorporation and inter-pretation of oxygen-18 and of water isotopes
• Add carbon-13 and carbon-14 to the CESM ocean, atmosphere, and coupler; (already included in land)
CRITICAL: Support from software engineersand the wider community
CESM - Challenges for an Earth System ModelContributions from the Paleoclimate WG
• Continue to apply the model for the quaternary and the deep past
• Complete three CMIP5/PMIP3 simulations: - last millennium (850 to 2005 transient)- middle Holocene (6000 years ago time slice)- Last Glacial Maximum (21,000 years ago time slice
• Collaborate with land ice WG on simulating past changes in ice sheets
• Continue to develop CESM/CCSM4 for deep time periods; need tools to handle prescribed aerosols, tidal mxing and ocean overflow
