"Economic models of climate change typically assume that changes occur gradually and reversibly," says Dr. Klaus Keller, assistant professor of geoscience, Penn State. "However, some environmental effects are not smooth and show a threshold response. For a long time nothing or very little happens and then suddenly a large change occurs."
Keller and William E. Easterling, professor of geography and director of the Penn State Institutes of the Environment, analyze two potential threshold responses to anthropogenic greenhouse gas emissions - a widespread bleaching of corals and a collapse of oceanic circulation systems. These events could happen suddenly and predictions about whether and when they would happen are uncertain.
These potential climate thresholds call into question the results of previous economic analysis of climate change policies, Keller told attendees today (Feb. 18) at the annual meeting of the American Association for the Advancement of Science, Washington, D.C.
"Economic analyses typically neglect that greenhouse gas emissions might trigger climate thresholds with potentially significant ecological and economic impacts," says Keller. "Analyses neglecting the potential fro negative impacts of greenhouse gas emissions are biased toward too high greenhouse gas emissions. We need more realistic representations of the potential environmental threshold responses in economic analyses."
The uncertainty about the climate thresholds and the potential for an abrupt response pose special challenges to the design of climate policies. One key question is whether the current observation system would deliver an actionable early warning signal about possible ocean circulation changes.
"Think about the situation where you are in a canoe on a river with a waterfall, says Keller. "You may want to know the location of the waterfall early enough to be able to avoid going over the waterfall. The situation for climate thresholds is similar. One may want to see early warning signals before it is too late to avoid the threshold response."
Key questions are how confident we have to be in an early warning signal before we consider it sufficient to take action, and how to design and implement an observation system that could deliver an actionable warning signal.
"Observation systems that would yield actionable early warning signal about climate thresholds have the potential to improve climate policies considerably. Implementing such observation systems could very well be a highly profitable investment for future generations."