Japanese recovery officials search through the ruins of the leveled city of Minamisanriku, in northeastern Japan, Tuesday March 15, 2011.

Japanese recovery officials search through the ruins of the leveled city of Minamisanriku, in northeastern Japan, Tuesday March 15, 2011. AP Photo/David Guttenfelder

Can the Military Design a Disaster-Resilient City?

Cities are incredibly complex, and break in complicated ways. Understanding that complexity will be key to mitigating tomorrow’s disasters.

There’s no such thing as a simple disaster. Take the 2011 tsunami that swept across  eastern Japan. As a wall of water moved toward low-lying towns and villages, people attempted to flee in cars — and drowned in massive traffic jams. The floods decimated infrastructure and roads, hindering rescue and recovery efforts. The water also swamped the backup generators of the Fukushima Daiichi nuclear power plant, leading to a Chernobyl-level meltdown. Planners might have predicted any one of those things individually, yet no one was prepared for what actually occurred.

The current “system of systems” approach to disaster-response planning basically piles up management practices for individual events — say, handling a blackout, managing event traffic, providing mass-casualty medical care, etc. — and generally produces, well, a big mess.  The Defense Department wants to improve on that method.

On Friday, the Defense Advanced Research Projects Agency, or DARPA, announced a new research initiative into the ways important things can break at once. Called the Complex Adaptive System Composition and Design Environment, or CASCADE, the project is meant to help planners make cities, towns, bases, power grids, etc. less vulnerable to devastation.

“It is difficult to model and currently impossible to systematically design such complex systems using state of the art tools, leading to inferior performance, unexpected problems, and weak resilience,” the agency wrote in a release.

Resilience is the key part. You can prevent terror attacks and you can prevent some disasters, but you can’t prevent all of them. However, if you can decrease the cost of the disaster in terms of lives disrupted, time lost, pain or the level inconvenience imposed on citizens, then disasters become less … disastrous. That’s what’s called building in resilience.

“We need new design and representation tools to ensure resilience of buildings, electricity, drinking water supply, health care, roads and sanitation when disaster strikes,” CASCADE program manager John Paschkewitz said in a statement. “CASCADE could help develop models that would provide civil authorities, first responders and assisting military commanders with the sequence and timing of critical actions they need to take for saving lives and restoring critical infrastructure. In the stress following a major disaster, models that could do that would be invaluable.”

Of course, urban planners have been trying to design away the costs of disaster since the 18th century BC, when Egyptian King Amenemhet III created one of ancient history's most celebrated examples of flood mitigation by using more than 200 wheels to divert the Nile’s rising waters away from farmland. Roman Emperor Augustus followed with the first professional fire department, the Corps of Vigiles, to keep small fires from engulfing all of Rome, a job formerly performed—poorly—by slaves.

Obviously, Amenemhet III and Augustus could not predict when a fire or flood would take place, only that disaster was inevitable, so they created policy based on unspecific but highly credible information: bad things will happen. Today, military and civil disaster response teams can receive live video footage of any disaster the moment it emerges, as long as locals’ smartphones can reach the web. In 2011, a few Silicon Valley researchers even patented a system to deliver customized evacuation routes to any individual in the event of a disaster, on the basis of traffic patterns, posts uploaded to social media, and other data.

Unlike Augustus, we’re awash in information. But how to actually use such data to make disasters less deadly remains an mostly unanswered question.