Methodologies which enable robust model-data comparison

Appropriate methods for model-data comparison must be used, taking into account the nature and sparsity of data, temporal and spatial correlations, likely scale-dependency in the performance of models, the question how to compare integrated temporal data to a snapshot of a climate model, the rates of changes in transient model-data comparisons, and the uncertainties in both model output and data. This theme includes topics and methods from spatial analysis (e.g. PCA, statistics of samples on a sphere), time-series analysis (e.g. wavelet transforms), the properties of qualitative versus quantitative data, and the development of robust, standardised methods for model-data comparison.

WikiDot Home Page

PEN - ESS Workshop

A PEN - ESS workshop: “Assessing Paleoclimate Uncertainty”, is being run jointly by the two Past Earth Network (PEN) working groups; Models and Data-Model Comparisons. The aim is to develop collaborative projects between the different communities. The programme will be a combination of lectures and discussion sessions, which we hope will generate both light and perhaps new ideas for collaboration. Visit the Workshop homepage here.

Working Group Leaders

The following members have kindly accepted to be the leaders for this working group:

Manfred Mudelsee - Climate Risk Analysis
Louise Sime - British Antarctic Survey
Richard Wilkinson - University of Sheffield

Working Group Funding

The PEN Data-Model Comparisons working group has a budget to fund travel, conference sessions, meetings and similar activities for working group members. The working group leaders welcome proposals for this funding which benefit the working group. Please join the working group mailing list to become a member and join the discussion.

Membership & Mailing List

If you would like to become a member of the PEN via the Data-Model Comparisons working group please use this link or send an email to staey@leeds.ac.uk with the Subject line "Data model comparison JISCMail request". This will also subscribe you to the Jiscmail mailing list.