All products featured on WIRED are independently selected by our editors. However, we may receive compensation from retailers and/or from purchases of products through these links.
Tipping points are found in ecosystems, economies and even bodies. But they're usually recognized in retrospect, when it's too late for anything but regret.
Now a growing body of research suggests there are telltale mathematical signals. If scientists can figure out how to detect them, they may be able to forecast tipping points ahead of time.
"We are repeatedly blindsided by disasters that come out of the blue. If we had better tools for anticipating those events, we could avoid some of them," said Steve Carpenter, a University of Washington ecologist and co-author of a review Wednesday in Nature.
In 1982, physicist Kenneth Wilson won a Nobel Prize for developing equations to describe transitions that don't happen in a linear, easily predictable way, but are sudden and massive, such as fluids becoming turbulent and metals becoming magnetized.
Since then, scientists have noticed similar shifts elsewhere. The theory provides the only models that make sense of the Sahara's sudden flip from fertile grassland to sandy wastes some 5,500 years ago. Exploited fish populations fluctuate wildly. Futures prices on the S&P 500 displayed telltale skewing in the year preceding the 1987 stock market crash.
The proposition is by no means certain, but the possibility of being able to predict these sorts of events is tantalizing.
"These are provocative possibilities. The fact that the patterns seem to recur in so many different circumstances suggests that the mechanisms underlying them may have universal characteristics," said study co-author George Sugihara, a nonlinear dynamics specialist at the Scripps Institution of Oceanography.
The math appears to be universal. Carpenter and others suspect that all critical transitions are preceded by the same basic patterns. The trick is to figure out what sort of data to look for, and then how to make sense of it.
Tip-offs for tipping points appear when the feedback loops that normally keep complex systems at equilibrium become stressed. Too many trees are cut down, too many cattle are turned out to graze, too many investors sell low. The system takes longer to recover from variations it normally weathers. Its mathematical representations become jagged rather than smooth.
*The data needed to detect ecological tipping points comes from Earth-orbiting satellites, but the U.S. fleet isn't up to the task. *
“Global and regional monitoring systems are a long way from where they need to be,” said University of Wisconsin ecologist Steve Carpenter. “Earth-observing capacity in the U.S. has deteriorated badly.”
*In 2007, a National Academy of Sciences report predicted the number of U.S. satellites would drop by 40 percent over the next three years. “There’s a train wreck coming,” said William Gail of Microsoft’s Virtual Earth. *
*In April, the National Oceanic and Atmospheric Administration said it needed $670 million to replace U.S. weather satellites, which will start shutting down by 2014. *
"The message to policymakers is that monitoring needs to improve a lot,"* Carpenter said. "** And it's not enough to go gather the data. Someone has to analyze it, someone has to pay attention to it. Right now, nobody is paying attention."*
Image: Colorado State University
Assuming scientists can find the signals that precede these events, and that monitoring systems can gather the necessary high-quality, long-term data — a very big if — researchers then need to figure how different kinds of feedback interact and where the thresholds points are.
In a rangeland, for instance, dozens of factors are involved in desertification, from fire intensity and grazing pressure to ambient moisture and soil composition. One might be more relevant than another to understanding the system's dynamics.
"It'd be very nice if it were true that there were precursors for tipping points in all these diverse systems. It'd be even nicer if we could find these precursors. I want to believe it, but I'm not sure I do," said Steven Strogatz, a Cornell University biomathematician who was not involved in the paper.
The difficulty of early detection is especially pronounced with markets. Computer models can replicate their bubble-and-crash behavior, but real markets — buffeted by political and social trends, and inevitably responding to the very act of prediction — are much cloudier.
"It is hard to find clear evidence of bifurcations and transitions, let alone find an early warning system to detect an upcoming crash," said Cars Homme, an economic theorist at the University of Amsterdam.
The most promising evidence of useful early warning signs comes from grasslands, coral reefs and lakes. Vegetation-pattern-based early warning signs have been documented in several regions, and transition theory is already being used to guide land use in parts of Australia.
The U.S. Geological Survey is currently hunting through satellite imagery for signals of impending desertification at two sites in the Southwest. They've studied desertification there by painstakingly measuring local conditions and experimentally setting fires, removing grasses and controlling the fall of water. But so far, the vegetation patterns that indicated tipping points in the Kalahari haven't shown up here, though this may be due to poor image quality rather than bad theory. The researchers are now looking for signals in on-the-ground measurements of vegetation changes.
"These things aren't going to be foolproof. There will be false positives and false negatives, and people need to be aware of that," said Carpenter. "There's still a great deal of basic research going on to understand the indicators better. We're still in the early days. But why not try? The alternative is to get repeatedly blindsided. The alternative is not appealing."
Citation: "Early-warning signals for critical transitions." By Marten Scheffer, Jordi Bascompte, William A. Brock, Victor Brovkin, Stephen R. Carpenter, Vasilis Dakos, Hermann Held, Egbert H. van Nes, Max Rietkerk & George Sugihara. Nature*, Vol. 461, No. 7260, September 2, 2009. *
Image: NASA. In March, researchers predicted that continued deforestation would cause less rain to fall in the Amazon, ultimately resulting in the rain forest's transformation to grasslands.
See Also:
- In the Bowels of Carnivorous Plants, a Tiny Model of the World
- Saving Fish is Possible, Unless They're Past the Tipping Point
- A Handy Guide to Climate Change Tipping Points**
Brandon Keim's Twitter stream and reportorial outtakes, Wired Science on Twitter.
