[an error occurred while processing this directive] [an error occurred while processing this directive]


Report 2

12 August 1997

 Chairs: John Kutzbach, University of Wisconsin and Bette Otto-Bliesner, NCAR



  The CSM Paleoclimate Working Group held its second meeting on June 25, 1997 in Breckenridge, Colorado during the Second CSM Workshop. The purpose of this meeting was to gather scientists interested in using the CSM and its component models for paleoclimate research. Letters of invitation for the Paleoclimate Working Group session were sent by e-mail to over 40 individuals at universities and laboratories working in either modeling or data studies of past climates. Participants in the working group meeting are listed in Appendix A.

  The meeting began with an overview by B. Otto-Bliesner and E. Brady (CGD/CCR) on the current status of the CSM for paleoclimate research. A slab ocean model based on the NCAR CSM Ocean Model (NCOM) was proposed by B. Otto-Bliesner. This model with one 50-m deep mixed layer and user-specified ocean heat transport is capable of being expanded to include features of the full ocean model if warranted. The successes and biases of a 130-year simulation with a low-resolution version (T31 atmosphere and land, x3 sea ice, x3 25 level ocean) of the full CSM were also discussed. The model results were favorably received. Concern was expressed over the weak El Niño signal produced by the CSM. Enhanced oceanic resolution in the tropics was suggested as a possible solution to this problem. Improvements in the component models to include better sea ice prediction, runoff, and interactive vegetation are particularly relevant to paleoclimate research.


  The Paleoclimate Working Group was charged with drafting a science proposal to be submitted to the Scientific Steering Committee (SSC). The group proposed four science thrusts that are of primary importance to the paleoclimate community and will benefit from the capability of the CSM to do "long coupled" simulations. A model testing activity crucial for future science thrusts was also identified. Three coupled simulations and some spinup experiments were identified as "ready to do" and feasible within the 1998 CSL resource allotment. Community meetings will be held near the end of the year to determine appropriate forcings, boundary conditions, analysis procedures, and data-model comparisons. Participants in these meetings will include a subcommittee of the Paleoclimate Working Group as well as members of other working groups that are interested in using the specific simulation.

 Climate of the 17th-18th-19th centuries

 The focus of this thrust is to make available a series of simulations which will allow a comprehensive evaluation of the natural (non-anthropogenically-forced) annual-decadal-centennial variability of the model to aid in the understanding of the detailed paleoclimatic database being compiled within the research community. The climate of the 20th century does not fully represent the true range of variability of the system in terms of extremes, persistence, and decadal to centennial variability. These simulations will allow evaluation of the patterns, ranges, and proposed causes of decadal-centennial variability and the coupling of the interlinked systems of ice, atmosphere, ocean, and biosphere in this response. Reconstructed time series of climatic forcing (solar variability, volcanic activity) for this time period should be available within the next year. The Decadal to Centennial Working Group has also expressed interest in using these simulations. The thrust will proceed in two phases:

Phase 1 (1998):

1) Baseline simulation with no anomalous external forcing

Model: T31 x 3 coupled CSM

Length of run: 1000 years

Phase 2 (1999+):

2)Baseline simulation plus additional external forcing from reconstructed solar variations

3)Baseline simulation plus additional external forcing from reconstructed volcanic activity

4) Experiment 2 plus 3

5) Experiment 3 extended into the 20th century and including anthropogenically-forced atmospheric CO2, aerosol loading, and land use.

Paleoclimate Modeling Intercomparison Project (PMIP)

As a corollary to AMIP, the Paleoclimate Modeling Intercomparison Project was developed in 1991 to coordinate a systematic study comparing simulations by atmospheric general circulation models of the large changes of climate that occurred in the distant past. PMIP is initially focusing on the mid-Holocene (6000 years before present) and the Last Glacial Maximum (21,000 years before present). PMIP simulations with CCM3 are currently being completed. Fully coupled CSM simulations for the PMIP time periods have been identified as a science thrust of the Paleoclimate Working Group. These simulations will provide an assessment of differences in variability during glacial and interglacial periods of altered climatic forcing, including such items as the characteristics of ENSO and other identified oscillations (NAO, Atlantic dipole, etc.), and role of the coupled response in accounting for unexplained biotic and hydrologic changes documented in the proxy data. The Decadal to Centennial Working Group will be collaborating on ENSO variability aspects of past climates.

Phase 1 (1998):

6000 and 21,000 yr. B.P. with CCM3

Models: T31 and T42; fixed SSTs and slab ocean

Length of runs: 10 yrs. fixed SSTs, 30 years slab ocean

6000 yr. B.P.

Model: T31 x 3 coupled CSM

Length of run: 500 years

Phase 2 (1999+):

3) 21,000 yr. B.P.

Glacial-Interglacial Climates and Abrupt Change

Analyses of ice, ocean, and lake cores have documented that the climate system has fluctuated dramatically on decadal to century time scales over the last 130,000 years. These signals in some cases appear to have a global signature. Meltwater impulses from the large Northern Hemisphere ice sheets as the climate warmed from the Last Glacial Maximum to the start of the Holocene (10,000 years ago) have been proposed as a possible causal mechanism of these major abrupt changes. Significant time lags suggest that decadal-centennial oscillations triggered by an impulse may have to be invoked to explain the record of change. Abrupt reorganizations during the Holocene have been hypothesized to be linked to Antarctic climate changes. Coupled CSM simulations will establish the modes of teleconnection between regions remote from the forcing. These simulations will be done in collaboration with oceanographers in the community.

Phase 1 (1998):

1) Idealized simulations exploring coupled sensitivity to Arctic and Antarctic meltwater impulses. (These simulations will probably include a small subgroup from the Paleoclimate and Ocean Working Groups but are important for Phase 2 of this thrust which will are of community interest.)

Model: T31 x 3 coupled CSM

Length of runs: 100-300 years

Phase 2 (1999+):

2) 8000 yr. B.P.: Early to mid-Holocene transition; proposed to be linked to Antarctic meltwater impulse

Younger Dryas (return to much colder conditions in the North Atlantic and Europe about 11,000 years ago)

4) Heinrich cycles (sediment layers in North Atlantic rich in ice-rafted debris and poor in foraminifera deposited at 5000-10,000 intervals)

Warm Climates of the Last 100 Million Years

Earlier periods of warmth are of great interest for the global change problem. Considerable work is ongoing in this area and should be encouraged. At the appropriate time, these studies may become the focus of detailed cooperative studies. Working group members will be monitoring these developments

Phase 2 (1999+):

1) Miocene: uplift, seaway closing

2) Mid-Pliocene: High CO2

3) Ice-free, extreme warmth time period

Ocean Spinup for Past Climates

The focus of this effort will be to explore the sensitivity of the coupled solution to the ocean spinup procedure. Experimentation with the CSM documents the importance of appropriate spinup of the ocean-alone model before coupling to the other CSM components. For this reason, Working Group coupled simulations for calendar year 1998 have been chosen such that the present-day ocean spinup was suitable for the proposed runs. Work will be done at NCAR to explore how to spinup the ocean when the model boundary conditions are significantly different than present. Specific projects identified for 1998 are:

Phase 1 (1998):

1) Present-day spinup with CCM3 rather than observed data

2) Sensitivity of spinup to initial conditions of ocean

3) Asynchronous coupling during spinup

Appendix A

Participants in Paleoclimate Working Group Session


Casper Amman

University of Massachusetts

Roger Y. Anderson

University of New Mexico

Steve Baum

Texas A&M

Esther Brady


Antonietta Capotondi


Robert DeConto

University of Colorado

Jan Dutton

Pennslyvania State University

Peter Fawcett

University of New Mexico

John Kutzbach

University of Wisconsin-Madison

Benjamin Felzer


Kirsten Menking

University of New Mexico

Carrie Morrill

University of California at Santa Cruz

Bob Oglesby


Bette Otto-Bliesner


Jonathan Overpeck


Judy Parrish

University of Arizona

David Pollard

Pennsylvania State University

Andy Robertson


Ed Schneider


Lisa Sloan

University of California at Santa Cruz

Karl Taylor

Lawrence Livermore National Labs

Starley Thompson



Individuals unable to attend session but expressing interest:


A. Broccoli


E. Pierazzo

University of Arizona