Empirical Dynamics: A New Paradigm for Understanding and Managing Species and Ecosystems in a Non-Stationary Nonlinear World

Principal Investigator: 
Proposal Abstract: 

Problem Statement:

  • There is a mismatch between the current ecological modeling paradigm based on stationary linear dynamics and nonlinear reality.
  • Unlike simple engineered systems, natural systems tend to be nonlinear, non-equilibrium and non-stationary. This is a key feature that applies now and 25 years from now.
  • The current ecological modeling paradigm based on stationary linear dynamics, is not predictive in real time (e.g., fisheries stock projections fail to satisfy this bottom-line).
  • Prediction is essential: for model validation; for detecting preeminent changes in state (regime shifts); and for “credible” exploration of plausible future environmental scenarios.
  • There is an important national need for better analytical tools and fundamental theory that realistically addresses natural non-engineered dynamic systems. Indeed, our ability to manage ecosystems 25 years from now will depend critically on having valid models that make credible forecasts.

Technical Objectives:

  • Develop “empirical dynamic modeling” (EDM) as a new conceptual framework that is predictive and explicitly accommodates nonlinear, non-equilibrium behavior.
  • Develop methods for incorporating stochastic elements and for quantifying uncertainty in EDM.
  • Expand EDM to accommodate multiple plausible futures that are non-analogue.
  • Test EDM by forecasting apparent threshold behavior evident in red tide events in Southern California.
  • Use the red tide model with future climate scenarios to investigate the prevalence of red tides in 25 years.
  • Extend EDM to high spatial, low temporal power data series common in ecosystem study.
  • Use spatial EDM to identify interactions among variables operating in Pacific coral reef ecosystems to provide guidance for management under climate change.
Start and End Date: 
September 2015 to September 2020
sharknado