Hacker News Reader

Comment by nkoren | Hacker News Reader

We should. I've seen an analysis which concluded that your chances of dying in a meteor impact are about the same as your chances of dying in a car crash -- eg., really quite high. This doesn't sound right, because at as an individual, you're obviously far more likely to die in a car crash -- it's extremely unlikely that you'll personally die from a meteor impact. However in the very rare event that you are killed by an impact, then there's a very high probability that anywhere from tens of thousands to tens of millions will also be killed by the same event. So overall the numbers roughly balance out, and are high enough to be worth taking seriously.

Fortunately there are groups that are doing just that: https://b612foundation.org/

View on HN

The main problem here is that detecting a city-killer meteor before it hits you is incredibly difficult, which really puts a damper on your ROI.

Meteors large enough to detect are both incredibly rare, and are very difficult to deal with, and meteors too small to detect will hit you before you see them.

Not to mention that any detection will come with very large uncertainty bars. Suppose that you somehow discover that a ~500 KT meteor (Like the Chelyabinsk bolide - which was only ~20 meters across - good luck figuring out an accurate orbit for it) is an impact risk. The overwhelming majority of the Earth is covered with water. The overwhelming majority of the rest is... Not densely populated. And observations put it at a 0.5% chance of striking the Earth next year, with zero ability to predict which part of the Earth will be struck.

That's the most likely outcome of a detection program. Not '100% chance', not '90% chance', but a '0.5% chance of a rock that's scary only if you're in its immediate path'. Is that worth doing anything about?

What's the cost of a nuclear missile (or X chosen method?) Whats the cost of NYC turning in a smoldering crater * 0.5% (or whatever)?

The cost of all rockets ever launched is a pittance compared to NYC real estate

Also there is a project for a space telescooe spesifically designed to locate asteroids, its not that difficult or even expensive.

https://en.m.wikipedia.org/wiki/Sentinel_Space_Telescope

Having nuclear missiles ready at all times, all over the world (near populated areas) is a risk on its own... In a politically stable world without conflicts, sure, but we haven't had that kind of world for centuries...

That's true, but we already have that. There's nukes all over the place. Not in all countries, but many.

> Not in all countries, but many.

In 9 out of 195 countries.

Why would they have to be stored near populated areas for this purpose?

Even if you were gonna hit the meteor during significant downward trajectory (which seems unlikely), that would still have to be done at quite a distance from the impact location.

Also wondering when there was ever a "politically stable world without conflicts"

Would even the largest nuke do much to a 50m-wide chunk of iron traveling at 20+ km/s?

Earth travels around the Sun at 30 km/s, delay an asteroid by 4 minutes and it will miss completely.

If your nuke can slow the asteroid from 20km/s to 19.9 km/s, then you need to nuke it 6 days before it hits earth.

The global reaction to COVID is highly relevant to this discussion. As with covid the secondary and tertiary effects will be vastly greater than the primary effects.

There is a fascinating logistics and timing feature to this whole topic. Aside from global panic'd masses. There is a huge logistical difference between sudden and unplanned electrical grid shutdown and subsequent black start vs a pre-planned five minutes outage (assuming we can predict impact to five minutes). Another interesting electrical utility issue is the radiological damage from a melted-down nuclear power plant varies exponentially with time... There's a huge difference in expected damage between whacking the coolant pumps a month in advance vs a day in advance vs a minute of warning vs no warning at all just no coolant flow.

About the water part: Historically you are completely right, but with the crazy amount of global shipping in the 21st century, I think lots of sea impacts would be detected and/or sink a couple of ships. Take a look at vessel finder and you'll see that there are only a few areas where ships are really far apart.

Reply on news.ycombinator.com