Prepping for Disaster

Prepping for Disaster

 Tom Easton

Most of these little essays deal with matters of science and technology. My aim has not been to add to the flood of techno-hype. I have mostly chosen to point out potential problems and—when possible—potential solutions. Perhaps unfortunately, the realm of solutions is not the realm of science and technology, but of politics. And politics is more complicated than science and technology.

At one point, I mentioned that some of the problems we face—global warming, overpopulation, and future food supply, for instance—are predictable. We can have some idea of when we will need solutions, and we can have the solutions ready. Or not (thanks to politics).

But some problems are less predictable. We have no idea when the next big earthquake or volcanic eruption will happen. We do, however, know that earthquakes and volcanic eruptions, as well as hurricanes and tsunamis, happen fairly often and can hammer large numbers of people. We have therefore invested in weather satellites and computer models that let us watch developing storms and predict their tracks. Early warning systems for earthquakes, volcanic eruptions, and tsunamis are less well developed and implemented. Indonesia’s tsunami warning system failed in December 2018, apparently because it was designed to listen for earthquakes (which often trigger tsunamis) rather than volcanic activity (

These disasters can’t be prevented. Money is spent on warning and monitoring systems because of their frequency and the damage they cause, both to lives and to property.  Also damage estimates can be based on past experience and compared with the cost of warning and monitoring systems. The cost-benefit analysis beloved of economists is both doable and favorable.

Cost-benefit analysis breaks down however, when the disasters are rare and large, chiefly because though the costs of early warning systems and even prevention programs are knowable, the benefits depend on what is known as the discount rate. To see how that works, consider a disaster a century from now that will do $1 billion worth of damage. If you invest $1 million today at 7% interest, the money will double every ten years and in a century you will have the billion you need to fix the future damage. Therefore you cannot justify spending more than $1 million today to prevent that future disaster. The interest you earn on that money is the discount rate. If the rate is worse or better than 7%, the money you can justify spending changes too.

It is at best awkward to try to put a dollar value on lives lost in a disaster, but most attempts to do so use numbers between one and ten million dollars per life ( A billion-dollar disaster can thus mean as few as 100 deaths, with no property damage. If the disaster involves millions of deaths, as well as extensive property damage, the bill approaches infinity and cost-benefit math just doesn’t work.

Are there any possible disasters that bad?

Consider the Yellowstone Caldera or supervolcano ( ). It last blew some 600,000 years ago, and volcanologists assure us that it isn’t about to do it again any time soon. But if it did, most of North America would see ash fall, and the global climate would be chilled, affecting food production.

It’s a maybe-someday sort of disaster. The cost would be huge, you can’t do anything to prevent it, and maybe suitable monitoring (future monitoring systems will surely be better than today’s) could see it coming. Given enough warning you could evacuate nearby populations (several states’ worth) and stockpile food. But how much can you justify spending today? And on what?

It would seem reasonable to spend money on better highways in the region to aid evacuation. They would also be useful in case of lesser eruptions in the region, and they would aid travel and commerce generally. A similar multi-benefit approach appears if you spend money on improved monitoring systems, for they can be valuable anywhere in the world where there is a volcano. Unfortunately, global benefits often don’t factor into political decisions.

Is there anything worse than a supervolcano?

Of course there is! Sixty five million years ago, a comet or asteroid 10 kilometers (6 miles) in diameter struck near what is now Chicxulub, Mexico: The results included the extinction of the dinosaurs (as well as a great many other species). Ancient history, you say? Meteor Crater in Arizona, almost a mile across, was created some 50,000 years ago by a meteorite 150 feet in diameter ( ). An even larger crater was recently discovered underneath the ice that covers most of Greenland; the date isn’t certain, but it may have happened only 13,000 years ago and caused the thousand-year-long global chill known as the Younger Dryas; see Paul Voosen, “Massive Crater under Greenland’s Ice Points to Climate-Altering Impact in the Time of Humans,” Science (November 14, 2018)(

Such events are rare. Chicxulub-scale events are very rare; a hundred million years may pass between them.  Meteor Crater-scale events may occur every thousand years, releasing as much energy as a 100-megaton nuclear bomb and destroying an area the size of a city.  A Greenland-sized event may happen every few million years; if one hit North America or Europe, it could destroy entire nations. If it hit in the middle of the Atlantic, the resulting tsunami would wash away entire seaboards.

How do you do a cost-benefit analysis on preventing disasters like these? You really do need to consider multi-benefit measures that can be justified in more than one way.

Warding off meteorite impacts is conceptually simple:


  1. Spot them on the way. This means using telescopes to inventory space rocks than come anywhere near Earth. According to NASA, there are some 4,700 asteroids more than 100 meters (330 feet) across, of which only 20-30 percent have actually been discovered so far. Greg Easterbrook, “The Sky Is Falling,” Atlantic (June 2008), argues that human society faces so much risk from asteroid and comet impacts that Congress should place a much higher priority on detecting potential impactors and devising ways to stop them.
  2. Develop ways of deflecting them so they miss Earth. Many methods are discussed in the “National Near-Earth Preparedness Strategy and Action Plan” prepared by the U.S. Interagency Working Group for Detecting and Mitigating the Impact of Earth-Bound Near-Earth Objects (NEOs) of the National Science and Technology Council (June 2018).


Most of these methods require a much more ambitious space program than anything the U.S or other countries have today. It would be immensely expensive, and no politician would be willing to try justifying it with a cost-benefit analysis. It would cost trillions, and in the worst case the damage estimate is very simple: Everything. Discount rate is meaningless.

Clark R. Chapman, in “What Will Happen When the Next Asteroid Strikes?” Astronomy (May 2011), argues that though the consequences of an impact would be catastrophic, efforts to prevent the impact would be futile.  Better, he says, to plan for evacuating target zones and providing shelter and food for the refugees.

We don’t have to buy that. A more ambitious space program would provide a great many benefits while we wait for an impact to threaten us. We could mine asteroids, build solar satellites to beam energy down to Earth, set up research bases on the Moon and Mars, and perhaps even—eventually—build colonies on other worlds.

Of course, supervolcanoes and cosmic impacts don’t pose threats any time soon. It’s all long-term stuff—though you really should keep an eye on NASA’s Coming Attractions list at . And politicians don’t think long-term.

Supervolcanoes and cosmic impacts do, however, become short-term concerns once we see them coming. With the first, we just need time enough to get out of the neighborhood. With the second… well, how much advance warning do we need to build up, in a great rush, a space program capable of warding off a big rock? Five years is surely not enough. Ten? Twenty-five? Fifty?

Surely, we need all the warning we can get. With no warning, disaster could come at any time. With not enough warning—that’s not much better.

It doesn’t matter how much it costs. It matters only that we can do it—in time.

Leave a Reply

Your email address will not be published. Required fields are marked *