Home | People | Research | Projects | School of Geographical Sciences

Spy glass

Palaeoclimate Modelling Intercomparison Project (PMIP)

People | Project Aim | Achievements | Future Planning | Scientific Themes | Meetings | Project Publications | Links


Co-ordinator: Sylvie Joussaume (LSCE)

Palaeo-environmental data set coordinator: Sandy Harrison (BRIDGE, University of Bristol)

Early holocene climate coordinator: Paul Valdes (BRIDGE, University of Bristol)

Project Aim

The aim of the PMIP project, which involves ca. 20 climate modelling groups and palaeoclimate diagnosticians worldwide, is to evaluate the robustness of climate model responses to external forcing. Modelling groups have performed parallel experiments in which the external forcing is identical. In the first phase of PMIP, there were two sets of experiments:

  • 6000 yr B.P. with a prescribed change in orbital parameters, in order to examine the simulated response to changed insolation.
  • 21,000 yr B.P. (LGM) with prescribed changes in geography (land/sea distribution and ice sheet extent and height, from Peltier 1991) and atmospheric carbon dioxide concentrations (LGM values of 200 ppmv). Some experiments were run with sea-surface temperatures and sea-ice extent prescribed from CLIMAP (1981); other experiments were made using a mixed-layer ocean.

Extensive model-model comparisons are currently being carried out to determine which features of the simulated climate are robust and which are model-dependent. Diagnostic analyses are helping to explain why some models yield different results from others at a regional scale. Comparisons with syntheses of palaeoclimatic data (see GLSDB, BIOME 6000, 21 ka tropical palaeoclimates synthesis) are being carried out in order to determine whether the simulated climates are realistic.

In the Second Phrase of PMIP (2002-present), the focus will be on evaluating coupled ocean-atmosphere and ocean-atmosphere-vegetation models. There will be four sets of experiments:

  • 6000yr B.P., with prescribed changes in orbital parameters and CO2
  • 21 000yr B.P., with prescribed changes in geography (land/sea distribution, ice sheet extent and height from the new ICE5G reconstruction).
  • 10 000yr B.P., with prescribed changes in orbital parameters, greenhouse gas concentrations and ice sheets.
  • 115 000yr B.P., to examine the processes underlying glacial inception.

More details of the experimental design are available from the PMIP website and Phase II PMIP2 website

Return to top


PMIP was designed as a model intercomparison project and initially focused on two periods, the Last Glacial Maximum (LGM, ca 21000 years ago) and the mid-Holocene (MH, ca 6000 years ago) that represent different forcing conditions. The LGM simulation was conceived as an experiment to examine the climate response to the presence of large ice sheets, cold oceans and low greenhouse gas concentrations. The MH simulation was designed to examine the climate response to a change in the seasonal and latitudinal distribution of incoming solar radiation (insolation) caused by known changes in orbital forcing. Seventeen modelling groups participated in simulations of these time periods with atmosphere-only models (PMIP1), and twelve groups in the second phase of the project using ocean-atmosphere or ocean-atmosphere-vegetation models (PMIP2). Several hundred scientists were involved in running and analysing the simulations, in producing palaeodata sets for model evaluation, and in model-model and model-data comparisons. PMIP has produced over 100 publications and PMIP results have been used extensively in the last two IPCC assessments.

Further details

Return to top

Future Planning

PMIP have identified 3 activity streams as the focus of their future plans:

(1) Benchmarking

Models that perform equally well for present-day and historical climate may produce very different responses to likely changes in forcing in the future. This makes it vital to evaluate and benchmark models, by comparing simulations of past climates against palaeo-observations. As new components addressing important feedbacks are incorporated within the current model framework, palaeo-benchmarking should be a critical part of the evaluation procedure. PMIP will take the lead here, by defining experimental protocols, assembling evaluation data sets, and undertaking quantitative assessments of simulations. Given the wealth of well-documented data sets already assembled, the LGM and MH will provide the focus for benchmarking activities within PMIP.

(2) Climate Analysis

The strength of PMIP lies in the ability to examine multi-model ensembles and to analyse the causes of differences in model ability to reproduce observed climate changes in the past. Thus, we will continue to focus on the analysis of the mechanisms of past climate change, specifically during past interglacials and warm periods, during intervals when there have been abrupt changes in the climate system, and during intervals when feedbacks related to changes in the land-surface or ocean circulation have played an important role.

(3) Exploration

PMIP has always provided a discussion forum which includes both modellers and observationalists, and is therefore extremely well placed to identify how emerging issues and uncertainties in global change science can be addressed through confronting models and data. There are clearly palaeo-dimensions to several emerging issues and uncertainties, for example, feedbacks through ice-sheet melting and sea-level rise, how vegetation changes influence trace gas and aerosol emissions to the atmosphere, the frequency of extreme events like tropical storms, and the relationships between changes in the mean state of the hydrological cycle and the occurrence of extreme floods and droughts. PMIP will therefore encourage the investigation of these issues and provide an active forum in which to discuss the results.

Return to top

Scientific Themes

Our future plans take into account the evolution of Earth system modelling, increased interest in climate variability and abrupt change, and the need to develop appropriate methodologies to understand and reduce uncertainties in climate projection. The work is divided into four major themes:

Theme 1: Evaluation of Earth system models (ESMs) at 6ka and 21 ka - Details

Theme 2: Interglacials and warm periods - Details

Theme 3: Abrupt climate change - Details

Theme 4: Uncertainties: characterisation and understanding - Details

Return to top


PMIP meetings

Autumn 2008, PMIP2 Workshop, USA
This workshop will focus on assessing the current state of PMIP analyses and on planning future work.
Spring 2010, Japan
This workshop will focus on the potential contribution of PMIP to AR5.
Summer 2011, Bern, Switzerland
This workshop will be linked to the INQUA Congress, and will thus provide an opportunity to focus on data syntheses and model evaluation.

PMIP involvement in externally-organised meetings

We already anticipate that there will be a significant PMIP presence at workshops and conferences organised by our sponsoring bodies and by other synergistic organizations, including:

April 2008, EGU
May 2008, IGBP Congress: PMIP will participate in a session on model-intercomparison projects, to ensure that model benchmarking against past climate changes is included in the forward planning of all model-intercomparison projects.
December 2008, AGU
July 2009, PAGES open science conference
2010, AIMES open science conference
December 2010, AGU: We envisage organising a session to present the results of ongoing PMIP analyses.

Possible joint workshops

(1) Reconstruction of interannual to interdecadal variability and climate extremes from the palaeo-record (jointly with PALCOMM and PAGES/CLIVAR)
(2) Reconstruction of the physical, biological and biogeochemical state of the 3-D ocean through the palaeo-record (jointly with PALCOMM)
(3) Past climate forcing (jointly with PAGES/CLIVAR)
(4) Pliocene climates (jointly with NASA GISS and USGS PRISM)
(5) PC4MIP (jointly with PAGES and AIMES)

Return to top

Project Publications

PMIP has produced over 100 publications and results from the project have been used extensively in the last two IPCC assessments.

Peer-Reviewed Papers

A selection of the most recent publications by PMIP members can be viewed here.

PMIP Flier 2008

The new PMIP flier summarises the current state of play of PMIP and looks forward to the future of the project. The flier is available in two versions, please open the one most suited to your needs:

  1. To view flier online only (non-printable version): download here!
  2. To print flier and view foldable version: download here!

Return to top


Main PMIP website