Pollen-Climate Methods Inter-comparison Project (PC-MIP) Workshop

13.06 - 15.06.2017  
Caux, Switzerland
Contact person:
Basil Davis, This email address is being protected from spambots. You need JavaScript enabled to view it.
Workshop report: 
> Access

The Pollen-Climate Methods Inter-comparison Project (PC-MIP) Workshop will be held in Caux, near Montreaux, Switzerland, from 13-15 June 2017.


The venue will be Villa Maria Seminar Centre.

Participation in the workshop will be open, but the total number of participants will be limited to 30. The Swiss National Science Foundation (SNSF) will support 10 international participants, with another 10 places for local participants from the Universities of Lausanne and Bern, leaving approximately 10 places free for open participation.

Timescales covered: The last 130,000 years.

Workshop goals

The main aim of the workshop is to undertake, for the first time, a comprehensive bench-test of the ~10 main numerical methods currently used to reconstruct climate from pollen data. From this analysis we hope to provide guidelines for use and recommend future improvements.

The workshop participants will undertake a series of experiments using pollen data from different locations and time periods chosen to help differentiate method performance. Sites will also be chosen to allow comparison with other proxies. Importantly, the evaluation process will not be based solely on statistical cross-validation using a modern pollen surface sample dataset, as has been commonly used in previous method evaluations. Many methods have been designed to improve the ability of the transfer function to address specific problems in the past, and not the present, such as low CO2 during the LGM or periods with poor modern analogues such as the Late Glacial and last Interglacial. We therefore plan to evaluate method performance over a range of different time periods over the last 130,000 years.

The main outputs for the PAGES community will be:

1) a high impact co-authored position paper in which we will present community-agreed recommendations for improvements in methods and guidelines for their interpretation,
2) enhancements to software used to make this type of analysis such as the 'rioja' multi-method R package, and
3) provision of reference data sets that will allow benchmarking of future methodological developments against previous methods.


Fossil pollen preserved in the sediments of lakes and bogs from around the world provides one of the main sources of proxy data for reconstructing past climate during the Quaternary period. Quantitative climate reconstructions from pollen data were first pioneered over 70 years ago (Iversen 1944). Since then the number of different methods has expanded greatly but with little co-ordination.

The objective of this workshop will be to bring together leading scientists in the field of pollen-climate reconstruction to undertake for the first time a comprehensive 'bench test' of all of the ~10 main numerical methods currently in use.

The workshop is particularly timely given new methodological innovations such as Bayesian statistics, and the rapidly increasing availability and quantity of pollen data thanks to public databases such as Neotoma. At the same time, we are also increasingly asking more from pollen data; seasonally resolved reconstructions, records of precipitation as well as temperature (even cloud cover), and always lower uncertainties and error estimates.

The workshop contributes to many themes within PAGES involving the use of proxy climate data, including activities Warmer Worlds, Data Stewardship and Thresholds, as well as the 2K, CVAS, DAPS, QUIGS and GPWG2 working groups. The workshop also has relevance for other proxies that rely on the same transfer function methods such as chironomids and foraminifera, as well as the use of pollen transfer functions to reconstruct land cover (LandCover6k working group).

Workshop outline (still to be confirmed)


1) Establish the experimental design, including choosing and preparing benchmarking pollen datasets (sites/time-slices),
2) Analyse these datasets according to the participants chosen methodology/expertise (where applicable),
3) Undertake outline planning of the manuscript, including background for each methodology: short history, reason for development, examples of applications/studies, what is good, what is not so good, known issues,
4) Discussion board for general issues and problems (eg. anthropogenic impact, climate parameters that can be reconstructed, site issues (bog/lake, marine, caves), calibration datasets (surface samples, maps of modern vegetation, climate data), rarefraction and taxanomic diversity, choosing key taxa in the assemblage and their taxanomic level, pft’s and biomisation etc).


1) Discussion of general issues and problems (see 4 above),
2) Presentation of the background for each method (see 3 above) and results of the pre-workshop analysis (see 2 above) (grouped by method),
3) General discussion of results and inter-method comparisons,
4) Summary and conclusions, including manuscript outline and assignment of responsibilities for its completion.


1) Undertake any additional experiments,
2) Complete manuscript,
3) Distribute benchmarking pollen datasets,
4) Complete PAGES report.

Confirmed key speakers

Pat Bartlein, Professor, Univ. Oregon, USA
Steve Juggins, Professor, Univ. Newcastle, UK
Simon Brewer, Professor, Univ. Utah, USA
Simon Goring, Senior Scientist, Univ. Wisc-Mad, USA
Heikki Seppa, Professor, Univ. Helsinki, Finland
Joël Guiot, Professor, CNRS/CEREGE, France
Richard Telford, Professor, Univ. Bergen, Sweden
Kenji Izumi, Postdoc, LMD/CNRS/UPMC, France
Manuel Chevalier, Postdoc, Univ. Montpellier, France
Norbert Kühl, Senior Scientist, Univ. Bonn, Germany
Lasse Holmstron, Professor, Univ. Helsinki, Finland

Applications and financial support requests

Application deadline is 28 February, but early application is encouraged. If you would like to attend the workshop, please contact Basil Davis (This email address is being protected from spambots. You need JavaScript enabled to view it.) with a short description of your motivation and research experience on the topic.

Participants will be expected to actively engage in both pre-workshop and post-workshop activities (organised remotely), as well as the workshop itself.

Financial support is available for a small number of participants, with a focus on early-career scientists and those from countries where funding is otherwise difficult to obtain. Please state in your application whether you would like to apply for PAGES support, and on what basis.

Further information

Contact meeting organizer Basil Davis: This email address is being protected from spambots. You need JavaScript enabled to view it.


Post-meeting material

pcmip 17 group web

Group photo from the PC-MIP workshop.


pcmip 17 discuss web

Group discussions during the workshop.


pcmip 17 adawson web

Andria Dawson discusses Bayesian approaches to paleoclimate modeling.