21,000 Years of Climate Change: Understanding Earth’s Past Through Models and Data

21,000 Years of Climate Change: Understanding Earth’s Past Through Models and Data 

How has Earth’s climate changed over thousands of years—and what can that tell us about the future?  This PhD project offers the chance to explore long-term climate shifts using innovative modelling tools and ancient environmental records. You’ll be part of a growing field that connects past climate data with modern science to better understand how our planet responds to change.  

Motivation 

Climate models are used to understand what the world might look like in the future. However, we need to know that they are providing accurate projections.   Validating the models against the instrumental record (since A.D. 1850) is useful, yet future climate change is projected to be far outside recent bounds. 

To provide a more rigorous test of climate models, we can look to palaeoclimates which were very different from today. However, there are no direct measurements of these ancient climates; instead, data from palaeoarchives (i.e. ice cores, speleothems, tree rings) are analysed to show what the climate was like. 

One important proxy derived from palaeoarchives is the water isotope ratio (d18O).  This provides an indication of temperature or precipitation which can be compared with climate model output. However, multiple processes influence the water isotope ratio, and it can be difficult to quantify what the water isotope signals represent.  Fortunately, water isotopes have been incorporated into some climate models, which allows for an enhanced understanding of past climates. 

This project will produce water isotope enabled climate simulations, for key intervals over the last 21,000 years. By directly comparing modelled isotopes with proxy records, the project will enhance our ability to evaluate model performance and reconstruct past climate dynamics. This will lead to a more accurate understanding of how climate has changed in the past and whether climate models are able to reproduce these changes. 

Methods, tools, and interesting questions 

You will incorporate water isotopes into pre-existing simulations of the last deglaciation (21ka-0ka) and investigate key timeslices. 

Firstly, you will look at the Last Glacial Maximum (21,000 years ago) and rigorously assess the model performance against palaeodata for this time.   The model can then be run forwards in time:  understanding how the melting of ice sheets, and changes in orbits and greenhouse gases will affect water isotope tracers and how these are related to climate change at different time periods and at different regions of the globe.   

Following initial tests, there will be the option of taking the project in different directions depending on your interests. For example, you could look at processes occurring over the deglaciation such as variability in the Atlantic Meridional Overturning Circulation (AMOC) or the El Nino Southern Oscillation (ENSO). Alternatively, you could look at rapid climate change events such as the Younger Dryas or the 8.2ka event, to see how they are represented in palaeodata from around the globe and whether the signals seen in the palaeodata are related to local temperature or precipitation changes or larger scale circulation features.  

Skills 

To analyse the climate model output, you will become skilled in data science, including statistical techniques, as there will be substantial amounts of climate model output and palaeodata to process and understand.  

You will also need to develop strong critical thinking skills and to question everything! There will be multiple uncertainties on model and data that you will need to account for, as well as strong climate feedbacks that may not be perfectly represented by the model. 

Because the project bridges the climate modelling and palaeodata communities, you will become adept at communicating across disciplines—translating ideas between researchers with different methodologies and knowledge bases. This will be essential for collaborative research and for contributing to broader scientific conversations. 

 Impact 

Although climate simulations incorporating water isotopes can be run, they have been under-utilized, due to the combined challenge of setting up climate simulations for palaeoclimates and additional complexities caused by the water isotope physics.  Therefore, there is a critical need for the additional information that this project will provide. 

This project will fill a key gap by providing robust model–data comparisons across major climate transitions. Results will be of great interest to both modellers (who want to understand how well their model works) and the palaeodata community (who want to understand what the signals in their data mean), providing greater synergies between modelling and data than are normally possible.