Forced and internal variability in temperature simulations and reconstructions of the common era

Resumen

Understanding climate variability and change, including recent anthropogenic warming, poses questions that cannot be answered based solely upon instrumental records. The Common Era (CE), and specifically the last millennium (LM), are immediate temporal intervals with comparable external radiative forcings as those of present day. Exploring the climate system blended response to the forcing conditions and to the mechanisms imposed by its internal dynamics during the CE, and particularly the LM, has the potential to expand our understanding of climate variability from inter-annual and decadal to multi-centennial timescales. This provides a wider context for current warming that might help constraining the uncertainty embedded in the future climate response to a sustained anthropogenic pressure. Proxy-based climate reconstructions, paleoclimate model simulations and estimations of external radiative forcing stand as fundamental tools that allow gaining insights about past climate variations, their amplitude and causes. The main aim of this Thesis is to address the relative roles of internal variability and external natural or anthropogenic induced changes in temperature reconstructions and model simulations of the CE. To this end, available continental, hemispherical and global temperature reconstructions, the ensemble of simulations including both Paleoclimate Modelling Intercomparison Project Phase III / Coupled Model Intercomparison Project Phase 5 (PMIP3/CMIP5) and non-PMIP3 model experiments, as well as the external forcing configurations applied are herein exhaustively compiled, analysed and intercompared, thereby providing a comprehensive overview of the current state of knowledge of the temperature evolution for this period. The relatively short ranges of external forcing variability within the CE/LM, compared to longer timescales like glacial/interglacial changes, for instance, represent a challenge in as much as the consistency between simulations and reconstructions can be affected by the large uncertainties in their respective responses to the external forcings. One of the core questions within this work relates therefore the extent to which a straight response to the external forcing can be identified during the period under study and whether this signal is common to simulated and reconstructed temperature. The comparison of simulations and reconstructions involves thus a suitable frame to identify robust features of past climate variations and also weaknesses that might affect the model simulations, the reconstructions or both. Within such a frame this work discuses the impact that a range of generally accepted methodological approaches might have on inferences about the consistency between models and the estimations from reconstructions, overall their influence on uncertainty. Also this work is aimed at responding to questions related to the influence of the spatial scales, from global to continental, on the temperature variations of the CE. The internal dynamics of the climate system, associated feedbacks and mechanisms gain a more significant role in as much as the regional scales reach prominence. Evidencing to what extent the role of the externally forced response in simulations and reconstructions suffers from the increasing presence of the internally driven temperature variations is one key concern of this Thesis.