Given a rough Earth analog that experienced a major nuclear war about 1 Ma, would there be any evidence of in the rocks in modern times?
All right, so I can think of two major potential lines of evidence off the top of my head when it comes to nuclear weapons:
1) Radioactive isotopes: Most of the radioactive isotopes in nuclear fallout are incredibly short lived, with half lives ranging from minutes to hours to days. (None of the common ones seem to have a half life that lasts more than a year.) So the blasted nuclear hellscape probably wouldn’t still be glowing in a million years, from what I’ve read. If nothing else, consider the fact that it’s safe for people to go to Hiroshima, Nagasaki, and the Trinity test site.
On the other hand, nuclear fallout does cause isotopic shifts that can be traced by chemists. For example, there are different calculations you have to do for pre-industrial and pre-nuclear samples in various kinds of radioactive dating (particularly carbon-14) because it’s caused the amounts of various atmospheric isotopes to shift. Strontium-90 levels also changed due to nuclear testing and that change is recorded in teeth, for example. However, 90Sr and 14C are both short-lived enough isotopes that I don’t think they’d be all that useful for the chemists in a million years. Presumably all the isotopes will have decayed away, though maybe there’s some magical chemistry that could be done looking at relative proportions of daughter isotopes. At this point we’re way outside my comfort zone; geochemistry was never my strong suit. But there is potential there, and if you want to go that route I’d suggest finding a geochemist to ask.
2) Sedimentary evidence. Probably more useful, if your future people have some geologists among them. If you had a worldwide nuclear holocaust, you’d end up with mass extinctions, large-scale fires, and presumably the collapse of civilization. So at the very least, your future explorers would find these signs. Paleontologists would see the evidence for mass extinction, and more damning, would potentially find massive boneyards in multiple locations all dating to the same time, that would indicate a single cataclysmic event. You’d also get charcoal layers associated with the extinction from worldwide fires, and occurrences of “nuclear glass” like the “Trinitite” found at the Trinity test site. All those could be geologically dated to the same time, which would be some pretty damning evidence.
Of course, since it’d be evidence preserved in rocks, they’d have to dig for it or find outcrops. But you should find that kind of stuff around. Heck, you could probably even find buried portions of cities (concrete is pretty hardy stuff; it’s already a rock) and maybe some shadows would be preserved on it, things like that. The trappings of civilization don’t necessarily weather away that fast, particularly not if they get buried in ash and sediment.
As a note, you’d see this kind of evidence preserved both on land (anywhere sediment is aggrading rather than eroding) and in the ocean. Ocean sediment cores would probably show some very strange things going on, an abrupt shift in sedimentation followed by a slow recovery.
Andrew later clarified that he was talking about a slushball Earth, with the global ice age touched off by the nuclear holocaust.
Now, I’m not entirely certain that a global nuclear war would set off global glaciation to begin with. I did some reading on the snowball Earth for a grad class, and I didn’t find most of the proposed mechanisms all that convincing other than lesser solar output and/or change in ocean circulation. The worldwide disaster from a nuclear war might throw a lot of particulates in to the air (and we know those will cause cooling) but they’ll fall out of the air fairly quickly, and consequently dirty up your snow.
But anyway. The slushball Earth isn’t something we need to debate here.
Even with global glaciation, you’ll still end up with geological evidence getting deposited in your oceans, even if at a different rate–but it’s something you’d be able to see with, say, a core drilled into ocean sediments. There’s a reason these kinds of corse get used often for paleoclimate research. In the slushball, there’s still open-ish water at the equator, which can allow for some sediment settling (such as say, the big ash layer) and input. Or if suddenly you’ve got what looks like normal sedimentation that has an ash layer than shifts to something odd like banded iron formations, that’s a big glaring clue that something weird and catastrophic happened.
Thoughts from other geologists?