P2C2: Bayesian Detection of Holocene Abrupt Transitions in Proxy Records and Climate Model Simulations

There is a lack of understanding of when and where abrupt climate events occurred in the past, and what are the drivers of such rapid climate shifts. This project seeks to identify the spatial and temporal components of abrupt climate transitions during the Holocene, the current interglacial period (~10.000-years). The researchers will combine paleoclimate data and models to assess the ability to simulate rapid climate shifts and predict future changes.

This project will leverage several recent developments, including a dramatic increase in the number of high-resolution Holocene paleoclimate records over the past decade, the development of new statistical techniques to detect changepoints in paleoclimate records (e.g., temperature or rainfall) and handle uncertainties of the paleoclimate data, and the availability of climate simulations with increasingly realistic drivers and processes.

The work aims to address three questions: (1) What is the spatial and temporal pattern of abrupt changes during the Holocene? Are they organized in time and space, or do they occur more stochastically? (2) What are the forcings (e.g., orbital forcing, solar variability, volcanic eruptions) and climate system nonlinearities (e.g., feedbacks, thresholds) that produce Holocene abrupt events? (3) Do coupled climate models simulate abrupt climate changes of the types observed in the Holocene paleoclimate record, or are models more stable?

To answer these question, researchers suggest to (1) use Bayesian changepoint detection to objectively identify rapid shifts in a large data set of Holocene proxy records while accounting for age, proxy, and methodological uncertainties, (2) aggregate these results into a systematic and integrated view of transitions in time and in space, (3) apply the same algorithm to several transient simulations of Holocene climate in order to test hypotheses about the relative roles of external forcing and stochastic internal variability in generating past abrupt events.

The potential Broader Impacts include a better characterization of abrupt climate events in the past, and an evaluation of their predictability with climate models. This project will generate curated paleoclimate data that will support the Findability, Accessibility, Interoperability, and Reusability (Fair) data management principles and federal data strategy.

The project will also provide training and support for one post-doctoral research and an undergraduate student. Additionally, undergraduate students will be involved through internship programs in Boulder, including the NOAA Hollings Program, the NOAA Educational Partnership Program with Minority Serving Institutions, and the UCAR Significant Opportunities in Atmospheric Science program.

The researchers will create new content for the “Paleo Perspective on Abrupt Climate Change” webpage hosted by the WDS-Paleo and NOAA’s National Centers for Environmental Information (NCEI), highlighting the findings from the project. This Paleo Perspectives are aimed at a general audience and explain how paleoclimatology data provide a long baseline of past change needed to understand the natural variability of the Earth’s climate over a variety of timescales. The perspectives also provide a variety of links to scientific research results and datasets.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.