Constructing a Model of Abrupt Climate Change

University Of North Carolina At Chapel Hill, Chapel Hill NC

Investigators

Abstract

This award supports the development and use of a mathematical model using coupled nonlinear differential equations to help explain the physical mechanisms underlying signals of abrupt climate change found in paleoclimate records. The LOgistic-Delayed nonlinear Differential Equation (LODE) model will be used to capture some important aspects of the physics of climate change at orbital and millennial scales, aspects of mid-Pleistocene climate, and Dansgaard-Oeschger oscillations of the last Ice Age. The LODE model possesses simplicity in its formulation (i.e., two equations, small number of adjustable parameters) and sufficient complexity in its dynamics (i.e., infinite-dimensional nonlinear delay differential equation) to simulate details of climate change that other simplified models cannot. The study will validate the LODE model, expand its scope, and computerize and disseminate the final version. Validation will involve comparing LODE's output with intermediate-complexity coupled ocean/ice sheet/atmosphere models that realistically treat ice cap evolution and greenhouse gas fluctuations. The validation process will include stability analyses to determine the sensitivity of the model to variations in the adjustable parameters, probe the limits of predictability, and determine the phase-space structure and bounds of the attractor. The final model is expected to be of mid-range intermediate complexity and include Earth's geometry and land/ocean distribution. It is planned that the model will be applied to data on centennial and decadal time scales. The research will provide a simple and user friendly mathematical model for the paleoclimate science research community that can be used help explain some aspects of abrupt climate signals extant in the paleoclimate record. The model can also be used as an important teaching tool for students in paleoclimate science. Furthermore, this award will provide support for a post-doctoral researcher working under the mentorship of the principal investigator.