Print PageEarth stewardship needs science of integrated social-ecological (or human-environmental) systems. To ecologists, humans are a perturbation; to non-ecologists, ecosystems are a source of food, recreation and natural disasters. If humans have a future, it is inseparable from the future of ecosystems. Yet narrowly disciplinary views dominate today, even though truly integrated social-ecological science is emerging. This talk focuses on one area of integration, research on extreme changes.
Alternate states are a familiar example of extreme changes. New states of social-ecological systems are emerging, perhaps at an accelerating pace, as human activity expands. Recognition of this phenomenon evokes research on thresholds and resilience. Because some state changes are disastrous, early warning indicators are useful. Recent research reveals statistical early warnings for some kinds of catastrophic change. Among these are thick-tailed probability distributions. In a thick-tailed distribution, the next record event is likely to be many times larger than the previous record. The expected loss from the next record flood is several times greater than the current record loss. Humans are more accustomed to thin-tailed distributions; if the record longevity for humans is 122 years, the person who breaks that record is much more likely to die at the age of 123 than the age of 244.
Thick-tailed distributions associated with regime shifts and natural disasters have non-intuitive characteristics. Thick tails can emerge from microcorrelations, tiny correlations among individuals that balloon as individual behavior is aggregated. Multivariate thick-tailed distributions concentrate dependency in the tails, so that extreme events in two or more drivers are unexpectedly likely and catastrophic.
Thick tailed distributions confound human judgment. Experimental psychology, economics and political science expose many ways that judgment fails at the tails. Yet global change expands thick-tailed risks that humans are poorly equipped to address. This is a fundamental challenge for social-ecological science that can only be addressed by an integrated approach.
Results/Conclusions
Until research catches up, a few rules of thumb exist for thick tails. Predictions are useful tools for research but have limited value for policy because the next important event will be outside the range of experience. Human dynamics driven by microcorrelations and social networks can polarize politics around controversial predictions, leading to gridlock. Research that exposes vulnerabilities, magnitudes of plausible impacts, and innovative response options is more likely to evoke adaptive change. Expect the unexpected, and prepare by building resilience of integrated social-ecological systems.
