WGCM invitation to participate
Request for "performance metrics"
Definition of model "mean climate"
IPCC deadlines and transfer of information to WG1 Lead Authors
WGCM invitation to participate in CMIP3 in preparation for the IPCC AR4
Request for "performance metrics"
Some of the IPCC WG1 lead authors (notably from Chapter 8 model evaluation) have requested specific information that might prove useful in preparing their chapters. The overall purpose of Chapter 8 of the AR4 is to assess the ability of the global climate models to make projections of future climate change. Many of the analyses you are performing will contribute to answering this question. In the Third Assessment Report, Table 8.2 included several ocean circulation diagnostics for each of the models, along with an observational estimate of these same quantities (e.g., Antarctic circumpolar current transport strength, Nino 3 SST standard deviation, etc.). For the AR4, plans include considerably expanding a table of this sort, providing characteristics of the mean climate state and additional metrics for quantifying the ability of models to simulate particular phenomena. The chapter 8 authors would like to rely on you to provide scalar measures of model performance for each model and, when practical, for the multi-model mean simulation. These measures might, as appropriate, be in the form of an RMS error, a correlation, a mean difference, a ratio of variances, or (as in the TAR Table 8.2) simply a scalar measure of some simulated and observed quantity. As one part of your analysis, therefore, it would be useful to synthesize your results into a single (or perhaps a couple of) scalar metrics that can be used to gauge how well each individual model simulates the aspect of climate that you are analyzing. This request should certainly not be considered a requirement, but if you can produce a scalar metric of this sort, the chapter 8 authors will be grateful.
A final note concerning metrics: For some quantities, observational uncertainty will be an issue. Whenever possible, therefore, an assessment should be made as to how this uncertainty might affect the apparent agreement or disagreement between models and observations. When more than one reliable observational data set is available, it would be helpful to calculate differences based also on the alternative.
Definition of model "mean climate"
The IPCC WG1 lead authors have also made the following suggestion meant to help unify the analysis of model results. In defining the 'mean climate state' of a model for comparison against observations there are number of choices that could be made, e.g. use model 'control runs' (which may have either preindustrial or present day trace gases), or use the '20th Century all forcings' runs (many of which are available as ensembles started from varying initial conditions). For the 20th Century integrations there is also a choice of meaning period. It is recognized that the optimal choice for a given problem may depend on a number of factors including the period over which observations are available, and the need for a non-drifting or non-trending model solution. We also recognize that some projects have already begun their analyses based on a particular choice. We therefore do not wish to prescribe a solution to this problem and leave it to the judgment of individual projects. However, in cases where there is a choice, we wish to encourage as much uniformity in the analysis as possible, and therefore propose that other things being equal, model mean climate is defined based on the 1981-2000 period of the 'all forcings 20th Century' runs (or the ensemble mean where appropriate).
IPCC deadlines and transfer of information to WG1 Lead Authors
To fit with the IPCC review process, papers will need to be submitted to peer-reviewed journals and available to IPCC lead authors by 1 MAY 2005 (REVISED TO 31 MAY 2005), and accepted for publication in a peer-reviewed journal by approximately 1 DEC 2005 (REVISED TO 15 DECEMBER 2005). These deadlines are set to allow lead authors time to consider all available literature in time for the first and second AR4 drafts. Electronic copies of papers need to be sent to the appropriate lead author contacts, with an electronic copy lodged with the WG1 Technical Support Unit (TSU) see below.
For the May deadline, the authors need some form of draft paper to assess. In almost all cases this will be in the form of a paper submitted to a peer-reviewed journal. An exception to the need for a peer-reviewed paper can be made for calculations of well-established diagnostics (e.g. where the diagnostic and the lines of inference from the results are described in a number of existing peer-reviewed papers). This would be at the discretion of the lead authors and would be subject to scrutiny by the WG1 Review Editors and Bureau. If you are planning work of this type I would advise establishing early direct contact with an appropriate lead author.
List of contacts for AR4 modelling chapters
Please email a pdf copy of your paper to the IPCC WG1 TSU (ipcc-wg1@al.noaa.gov) and to the relevant contact points below. Note that your study may be relevant to more than one chapter (e.g. a study that considers both the control simulations of ENSO and future changes in ENSO would be relevant to both Chapters 8 and 10).
Roughly speaking, Ch. 8 deals with simulations of aspects of the climatology or processes in present-day climate, Ch. 9 addresses simulation of trends or responses to specific forcings in 20th century climate in addition to detection/attribution studies, Ch. 10 assesses any aspect of climate change projections from global models, and Ch. 11 deals with aspects of regional climate change projections from regional models or information downscaled from global models. These are not strict boundaries, but are used for general guidance to give you a rough idea of where to send copies of your papers. As long as you send copies to the chapter contacts you think are relevant for your paper, it is then the responsibility of the respective lead authors to sort out exactly where your results will be assessed.
Chapter outlines can be found at: http://ipcc-wg1.ucar.edu/index.html
Chapter 8: Climate Models and their Evaluation
Please send copies to the appropriate section contacts:
8.1 Introduction and Philosophy:
Richard Wood
richard.wood@metoffice.gov.uk
8.2 Advances in Modelling (model formulation, resolution etc):
Akimasa Sumi
sumi@ccsr.u-tokyo.ac.jp
8.3 Evaluation of contemporary climate as simulated by coupled global models:
Karl Taylor
taylor13@llnl.gov
8.4 Evaluation of large-scale climate variability:
John Fyfe
John.Fyfe@ec.gc.ca
8.5 Model simulations of extremes:
Jagadish Shukla
shukla@cola.iges.org
8.6 Climate Sensitivity and hydrological sensitivity:
Sandrine Bony
Sandrine.Bony@lmd.jussieu.fr
8.7 Mechanisms producing thresholds and abrupt climate change:
Ron Stouffer
Ronald.Stouffer@noaa.gov
8.8 Representing the global system with simpler models:
Thierry Fichefet
fichefet@astr.ucl.ac.be
Chapter 9 : Understanding and Attributing Climate Change
Please send all papers to CLA Gabi Hegerl
hegerl@duke.edu
Chapter 10: Global Climate Projections
Please send all papers to CLA Thomas Stocker
stocker@climate.unibe.ch
Chapter 11: Regional Climate Projections
Please send all papers to CLA Jens Christensen
jhc@dmi.dk
The draft papers will be made available through the WG1 web pages to all WG1
lead authors. Additionally, the TSU will set up a tracking sheet, and it will be
your responsibility to update this form as your paper progresses through the
review process. Further details regarding this procedure will be forthcoming.