Stanford Researchers Devise Method to Predict Black Swan Events

Researchers from Stanford University have developed a method for forecasting black swan events. Black swans are rare, unprecedented events with enormous consequences.

In tackling the supposedly unpredictable type of event, the bioengineers behind the project took an "unconventional approach by bridging the fields of statistical physics and ecological forecasting." Theories from avalanche systems were applied in long-term monitoring data from natural ecosystems. Avalanche behaviors refer to short-term, usually unprecedented, and unexpected behavior.

In a recent report, Samuel Bray recalled his observation in his study of microbial communities. Bray is a research assistant to Bo Wang, an assistant professor at Stanford's bioengineering department. There were instances where one of the species went through an unexpected increase in population far ahead of other species. He discussed it with Wang, and together, they asked whether this occurred outside the lab and if it is predictable. It led them to datasets on ecosystems closely monitored by the science community for years.

French Riviera Masters - Day Two
PROVENCHERES-SUR-FAVE, FRANCE - SEPTEMBER 22: A black swan besides the 6th green during the second round of the French Riviera Masters played over the Chateau Course, Terre Blanche Resort on September 22, 2012 in Provencheres-sur-Fave, France. Photo by Phil Inglis/Getty Images


They later settled with the three ecosystems. One was an eight-year study of Baltic Sea planktons, with the species levels measured twice a week.

The second dataset contained net carbon measurements of a deciduous forest at Harvard University. The data gathered every 30 minutes for almost 30 years.

The last one concerns barnacles, algae, and mussels measurements taken from New Zealand for over 20 years.


"It connects us to the bigger world."

"By analyzing long-term data from three ecosystems, we were able to show that fluctuations that happen in different biological species are statistically the same across different ecosystems," Bray said.

He added that their work "suggests that there are certain underlying universal processes" to take advantage of in the hopes of forecasting these rare but consequential events.

Wang, who is also the senior author of the paper, described their work as "exciting," pointing out the opportunity to take the computation tools they are developing to better understand or even forecast what will happen in the world.

"It connects us to the bigger world," Wang said.


The Black Swan Theory

The term "black swan events" was first coined in 2007 in a book of the same name by Nassim Nicholas Taleb, a Lebanese-American mathematical statistician. In his book, Taleb explained that these occurrences gave a disproportionate role in terms of magnitude and consequence. He also argued about the supposed non-computability using scientific methods, noting the nature of minimal probabilities. He also explained whether an event is considered a black swan occurrence affected by the observer.

Some black swan events in ecology might include the rapid decline and eventual extinction of passenger pigeons, once the most abundant species of pigeon endemic in North America. Passenger pigeons went extinct at the start of the 20th century.

A more recent example occurred with the death of two-thirds of all saiga antelopes back in 2015. Researchers in central Kazakhstan witnessed more than 200,000 saiga antelopes losing balance and dying shortly after--all in only three weeks.

READ MORE: Thousands of Endangered Antelope Deaths Have Scientists Investigating a Serious Concern

Another example is the 2008 global financial crisis, described by financial institutions as the most severe economic meltdown since the Great Depression at the onset of the 1930s.

Join the Discussion

Recommended Stories

Real Time Analytics