Project 1: Causal interactions in deep time
It is a quirk of human cognition that we tend to process temporally extended, dynamic information by segmenting it into discrete events with a beginning and an end. Longer-term patterns are often perceived as a background, or status quo, against which events are identified and often retrofitted with a cause-effect structure. In geology, this event-based causal focus is enhanced by the discontinuous nature of the stratigraphic record (which, “like the life of a soldier, consists of long periods of boredom and short periods of terror” in Derek Ager’s famous aphorism). The geological time scale itself is scaffolded by distinct events that allow chronostratigraphic correlation. Reasoning about long-term dynamics demands a conception of causality that goes beyond event-based characterization.
Geological records enable us to observe only a very small subset of the detail complexity (the number of interacting components) of the past Earth system. In principle, however, deep-time records can capture important aspects of its dynamical complexity, a concept which implies that cause-effect relationships may be far from obvious. In this project, we confront geological time series with statistical techniques designed to detect drive-response relationships in complex, nonlinear dynamical systems.
Examples of geological records of interacting Earth system components over the last 500 million years. The inferred information flow between the components is from a study by Hannisdal and Peters 2011.