Palaeoclimate modelling of monsoons during past warm periods

• Main Author: Zhao, Anni
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: UCL (University College London) 2023
• Subjects: Arctic; Global warming
• Online Access: https://discovery.ucl.ac.uk/id/eprint/10166788/2/Thesis_Anni_Zhao.pdf; https://discovery.ucl.ac.uk/id/eprint/10166788/

Three past warm periods were chosen to provide out-of-sample tests for those state-of-the-art climate models by phase 4 of the Palaeoclimate Model Intercomparison Project (PMIP4): the mid-Holocene (6,000 years ago), the Last Interglacial (more precisely 127,000 years ago) and the mid-Pliocene Warm Period (roughly 3.2 million years ago). Experiments were designed for each warm period with improved boundary conditions and protocols; called midHolocene, lig127k and midPliocene- eoi400 respectively. This work looks at the monsoon behaviour across the three PMIP4 experiments for the first time, to improve the understanding of palaeomonsoon and to evaluate the performance of current state-of-art models. Results of this work indicate that both the orbit-induced experiments (midHolocene and lig127k) show enhanced monsoons in the Northern Hemisphere and weakened monsoons in the Southern Hemisphere as expected. The lig127k simulations have stronger response than the midHolocene, because of their stronger orbital forcings. Simulated anomalies are generally in good agreement with climate proxy reconstructions, but both experiments underestimate the amplification of the northern African monsoon as well as Arctic warming. The midPliocene-eoi400 simulations indicate a global warming with a clear pattern of polar amplification, wetter tropics, and enhanced monsoons but with uncertainties. An idealised aerosol experiment highlights the potential importance of uncertainty in the aerosol specifications in the experiment protocol to simulating the mPWP climate. Analyses on the data-model mismatch highlight the source and importance of uncertainties during different time periods. Despite the existing uncertainties in the simulations, the results of the three experiments are useful for understanding climate response and quantitatively evaluating model performance. The findings from this thesis, combined with future work, improve our understanding of monsoon forced responses and could help to ensure that the next generation of climate ...