Arctic climate variability during the past millennium : combining model simulations and proxy data

• Main Author: Crespin, Elisabeth
• Other Authors: UCL - SST/ELI/ELIC - Earth & Climate, UCL - Faculté des Sciences, Fichefet, Thierry, Goosse, Hugues, Crucifix, Michel, Widmann, Martin, Swingedouw, Didier, De Keersmaecker, Marie-Laurence
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: 2014
• Subjects: Arctic
• Online Access: http://hdl.handle.net/2078.1/150597

The Arctic is one of the regions that experienced the fastest and highest warming of the planet during the recent decades, and changes in its environment had worldwide implications. Studying climate changes during the past is essential for a comprehensive understanding of the climate system, particularly to provide insights into the processes involved in the recent Arctic warming, in order to assess future global climate changes. Information about past climate evolution can be extracted from “proxy†data, while climate models can be used to interpret the observed changes. In this doctoral thesis, we have taken advantage of the complementarity of model results and proxy data, through data assimilation, to provide reliable simulations of the Arctic climate over the past millennium that are in agreement with all the possible sources of information: proxies, physics of the model and external forcings. Those simulations have then been used to analyse the role played by different forcing elements in the past Arctic climate changes. We have highlighted, for instance, the dominant contribution of volcanic eruptions in the simulated temperature trends and of the astronomical forcing in the differences in seasonal trends. The role of the atmospheric circulation in explaining some warm periods has been underlined in our simulations assimilating sparse temperature reconstructions based on proxy data. We have also demonstrated the need for a precise estimation of proxy data uncertainties to avoid inconsistencies between those data and the model physics. This study, being among the first to apply a data assimilation method in palaeoclimatology, provides a powerful tool for investigating past climate and some challenging perspectives. (SC - Sciences) -- UCL, 2014