It's a lot more difficult to launch through a rock ceiling, radioactive atmosphere, and a cloud of Kessler Syndrome fragments, from a gravity well 1G deep, than it is to launch from the surface of a rock with barely any atmosphere, a bare handful of satellites, in a well only 0.38G deep.
Also, bear in mind that you won't just be stockpiling food, but sufficient biodiversity to reassemble a viable ecosystem, and sufficient industry and education to preserve a viable economy. Otherwise, you still die when the nutri-mat and atmo-pure machines finally break down.
It's certainly easier to dig a deep hole and build a bunker in it, but that expense does not advance the end goal. It is merely insurance against a temporary setback. You still need to establish a launch-capable society on another planet, or everything dies anyway when the sun expands.
Then, after that happens, someone will still be around to say, "It would be cheaper and easier to colonize and terraform Venus than to send generation ships every which way."....
But isn't that what colonizing Mars is, anyway? Mars has no biodiversity to speak of and won't be a host to ours without significant evolution or resources, so what's the plan there?
Earth isn't just a convenient place to live. It's the only planet we can live on. Until we get warp drive or are much farther along in making a profit colonizing the solar system, not to mention in building better governments, it's just far too early to discuss colonizing Mars. Dreams are great but let's not get blinded by the hopes here. Let's plan a visit, and plan another visit 50 years after that.
Ugh, don't get me started there. Musk wanting to do all these internet satellites interests me but seriously gives me nightmares too in regards to Kessler syndrome.
Maybe someone launches anti-ballistic-missile missiles, and scores a few hits. Debris is spreading to all kinds of orbits. Kessler Syndrome will be the least of our problems as the bombs fall, but it'll be a pain later, after the fallout clears.
Sure, an extra backup on Mars helps... but let's keep the backups we can do now, first.
Also: spaceships... and tunnels. Next Elon enterprises should include underwater, Antartica or mountain-top habitats... and good practice for a Mars colony.
I wouldn't pay through the nose for that, either. Especially when we could be putting that time and energy into not losing the PROD deployment, today, with a much better ROI.
We don't need to worry about hypothetical 50-million year meteor strike extinction events. We're in the middle of one, that we could be taking action to deal with - for much less the cost of a self-sufficient Mars base.
You're worried about off-site backups, when the trash can in the data center is on fire, and you cant budget for a fire extinguisher.
There are certain types of disasters you can't protect against though, no matter how much money you spend. The only thing you can do is mitigate.
https://www.space.com/41080-alien-life-may-be-rare-today.htm...
So colonising somewhere other then Earth should also be a priority. A big risk are meteors. Just like the dinosaurs.
No they aren't. ~1km wide asteroids impact the Earth on average every 500,000 years and even then, it's too small to be an extinction level event.
Not to mention all NEOs around that size are constantly monitored.
Though the direct effects would be mostly local, the systemic effects would reverberate globally since in the modern world things (such as food production, supply chains, etc.) are so meshed together.
Draw a 300 km circle at your chosen impact point, then say bye to everything inside the first 150 km. The rest of the circle would be damaged to a varying degree. You can easily find places for impact points which, although not an ELE, would still turn many, many people's lives nasty.
Even if such a "smallish" 1 km asteroid hits, in general everything will change. For example, the asteroid could still cause enough chaos to weaken a country or countries, opening up for opportunistic wars. Since the long-term effects of a non-ELE would be transient, some countries might want to take advantage of this situation and invade (resource grab or glory for motherland or righting old wrongs or something). And then, what things would be seen next - for this we would be in unknown territory.
Even without an outright war, the number of displaced people could become very large. People would likely not sit still and clean up, but try to move away. Can the receiving end handle that many people, can they move enough help in when the just-in-time supply chains are disrupted to pieces? If not, there will be a collapse into some form of chaos at the receiver. And again, people might move, etc.
For an ELE, an offworld base is the only option for continuity. If that is not in place when an ELE hits, then humankind will be dead.
A Mars/Moon base operating at large enough industrial and R&D capacity (akin to a small to medium sized country) could help restore normalcy even without an ELE.
So, I think an offworld base is required anyway, even if we don't sweat so much about 1 km rocks, but still want to be serious about protection against an ELE.
It's estimated we've only found 20-30% of these objects.
A self-sufficient offworld colony is the best bet for normalcy and sounds like an overall faster and workable way for bootstrapping things again. Also, the technology developed as a precursor for offworld colonies would likely aid in things like detecting and diverting asteroids.
Colonizing the difficult places on the Earth, such as seabeds, sea bottoms and places deep underground, would be a good way to practise for the everyday life offworld (can't go out just like that, all food farmed indoors, strict trash recycling regimes, etc.) but I don't think it would be equivalent as a "backup".
It is, unless the scenario you are hoping to survive is "Hordes of proles at the gates."
https://www.space.com/9683-surface-mars-possibly-shaped-plat...
and
https://www.space.com/17087-mars-surface-marsquakes-plate-te...
A few hundreds of meters down and the nuke risk is negligible. You can stockpile enough for hundreds of years at ~0 shipping cost. You can set up fortified ducts/purification systems to get oxygen from the surface -- with backup allowing a small population to live long enough for the surface to get tolerable.
Heck, even a full blown self-sustaining underground habitat doesn't seem out of question, since there's water readily available, plenty of perhaps geothermal or nuclear power power, and construction minerals. Destroying the Mars habitat might not be too much more difficult comparing to destroying this colony. If it has less than 10,000 or so colonists the US gov. or similar might even have a chance of keeping its location in secrecy.
Sure it doesn't cover all possible scenarios, but in terms of backup a conflict-resistant habitat here on Earth seems like the best bang for your buck.
Welcome to Zion.