Paleoclimate data are influenced by measurement and proxy errors as well as noise generated within the climate system. In addition, paleoclimate time series often exhibit uneven time spacing, hiatuses and dating uncertainties, while spatial paleoclimate data often have non-uniform coverage with occasionally large gaps. Obtaining quantitative information with realistic error bars from such data is therefore a methodical challenge. However, paleoclimate data analysis is not only of genuine interest, it may also help us to put current climate changes into context.

We welcome contributions that take up the challenge and exemplify how modern statistical methods (Bayesian, bootstrap, Monte Carlo simulations) can help to aggregate and analyse uncertain paleoclimate data. Examples include, but are certainly not limited to, reconstructing temperature fields for the past millennium, Last Glacial Maximum or Pliocene, comparing global signatures of Dansgaard-Oeschger events, developing time slices for Cenozoic climate evolution or generating profiles of oceanographic parameters. We especially encourage contributions focusing on quantitative data-model comparison.

Confirmed invited contributions:

Martin Tingley, Harvard, USA; Masa Kageyama, CNRS, France