The best example for what I am trying to describe would be the worlds of Adventure Time, and The Dark Tower series.

These worlds both developed following a cataclysmic war that destroyed a relatively modern/advanced civilization, after which the by-products of the war (basically radiation, but it's technically not radiation so not bound by those rules) mutated the surviving flora & fauna as the world recovered.

I'm currently working on a world similar to these, with the apocalypse occuring in the late 1980's/early 1990's. In this world, the planet has more or less fully recovered. Although there are sure signs of the apocalypse, ecosystems like forests and other lush environments once again exist.

My question is how long would it take for the world to recover following an apocalyptic event that wiped out 90% of the worlds plant/animal life?

I will define "recovering" as the reappearance of robust and diverse eco-systems regardless of whether or not those ecosystems are even remotely similar to the pre-apocalypse. As far as human habitability goes, just assume that in these recovered/mutated worlds there are edible flora/fauna.

The type of destruction would be caused by war, so the scale of destruction would be consistent with all-out nuclear warfare preceded by years of conventional global conflict.

To address the comment concerning what the remaining 10% of life consists of/ if it is concentrated in one location or not:

I would say that the remaining 10% of life consists of random populations of certain random species scattered all over that just happened to be lucky enough to survive the main cataclysmic event of this apocalypse because of geographic protection, as well as species that are simply resilient enough to survive (No getting rid of those tardigrades)

EDIT: I specified that the "radiation" was not technically radiation because my world isn't constrained by the mutagenic effects of real life radiation, however I'm realizing that doesn't help anybody answer my question.

When it comes to contamination just consider it normal radiation/nuclear waste.

  • 5
    $\begingroup$ If a dinosaur can observe a post-Cretaceous world, he would definitely think it's posapocalyptic. It's been over 60 million years, and we still haven't recovered... $\endgroup$
    – Alexander
    Sep 12, 2017 at 17:58
  • 2
    $\begingroup$ @other comments: I don't think one can compare this scenario with other mass extinctions. Those were significant changes to the ecosystem stretched over a couple of (million) years where almost all species vanished. 90% of plants and animals dying overnight - earth might recover mostly within a couple of years and maybe completely in a couple of decades unless this is very selective, wiping out important species. Some systems might be hit a bit harder (forests, tundra), but over all ... How much % of plants and animals die each winter? $\endgroup$
    – Raditz_35
    Sep 12, 2017 at 18:04
  • 4
    $\begingroup$ It really depends on what you mean by '$90\%$' Is this remaining $10\%$ scattered all over the world or concentrated on one remote island? Is it made up of one species which was resilient but all of the others were entirely wiped out? $\endgroup$ Sep 12, 2017 at 18:10
  • 2
    $\begingroup$ How would you describe "recover"? To what extent the earth should recover until it satisfy your definition of "recover"? Wait, there's an answer asking for this :) $\endgroup$
    – Vylix
    Sep 12, 2017 at 18:17
  • $\begingroup$ I think you should define your apocalypse. Radiation that is not a radiation is not helping. If it was a nuclear war, for example, you will have to deal with the nuclear waste, and it requires a substantial amount of time to decay until it become "harmless" $\endgroup$
    – Vylix
    Sep 12, 2017 at 18:28

3 Answers 3


Let's start with the ugly stuff.

Nuclear fission produces the really ugly radition, but it goes away quickly. Fusion doesn't produce anywhere near the same fallout radiation, but what's left over takes forever to go away. My source suggests the fission problems follow the "rule of 7." Every 7 days the radiation drops 90%. From this perspective, the surface is "radiation free" (or, basically the way we live today) in less than 100 years.

But that's not the problem.

Fusion produces (listen to me sound like I know what I'm talking about. My thanks to that source!) ...produces tritium and C-14 carbon. These can be absorbed into DNA and therefore cause mutations (usually death-dealing cancer, but who knows, there might be a spider-man out there!). Tritium's half-life is about 12 years, but C-14's half-life if almost 6,000 years. That means the plants are mutating... and anything eating the plants are mutating... for six millenia. Ugh.

But that's not the problem.

The problem is the consequence of being forced to adapt to those circumstances. 6,000 years isn't enough time to evolve, only to adapt. No new species, but maybe improved species. At least, "improved" in terms of radioactive tolerance. The problem is that once DNA is affected, it can take forever and a day to weed out the genetic (and potentially hereditary) anomolies. The radioactive carbon is gone in 6,000 years (no new mutations! Yay!) but the genetic defects will take who knows how long to go away.

That's the problem, the time it takes real and honest evolution to work out the genetically weak and leave only the genetically strong. That could be a long honkin' time.

  • $\begingroup$ "That means the plants are mutating... and anything eating the plants are mutating... for six millenia". No. First, radiation causes radiation damage, not viable mutations that persist. "Radiation causes horrible mutations" is an overused and factually inaccurate. Also... half-life is the time it takes to half the radiation, not dissipate it completely. So when you say "The radioactive carbon is gone in 6,000 years" you are entirely wrong. $\endgroup$
    – MichaelK
    Oct 12, 2017 at 8:18
  • $\begingroup$ Also, radioactive intensity is inversely proportional to half-life. Long half-life = low intensity. This means that the amount of C-14 you need to cause damage multiplies with the half-life, not the other way around. Also.... you are already radioactive from C-14.... feel yourself mutating yet? $\endgroup$
    – MichaelK
    Oct 12, 2017 at 8:20
  • $\begingroup$ You wrote “My source suggests [...] every 7 days the radiation drops 90%”. Nope. Your source says the radiation drops 90% when you multiply the time since the explosion by 7. “multiply by 7” is definitely not the same as “add 7 days”. This is not the kind of exponential decay you get with a single nuclide. It's a power-law decay (as t^(−1.18)), which is explained by the fact that you have many nuclides with vastly different half-lives. $\endgroup$ Oct 12, 2017 at 9:26

enter image description here

As another person mentioned, what in your mind constitutes recovery?

Using Chernobyl as an example, nature is taking back the area around the damaged power plant rather quickly. In the decades since the accident, forests have sprung up regardless of former human activity. Even as it takes back, it doesn't completely heal itself. The buildings and traces of man are still there, but every season nature takes back more and more. But how long is long?

Using an old growth conifer forest as an example, if you log an area within the forest, it will take at least 600 years for the forest to make that clearing indistiguishable from the other parts of that ecosystem.

Damage done to areas with permafrost take even longer to heal. Roads and paths abandoned for over 100 years will still be easily identifiable.

During the last big extinction event at the end of the Permian–Triassic period, 90% of life on earth was wiped out. This included insects, plants, marine animals, amphibians, and reptiles. Life that was on the earth bounced back in a few years. Anything that could capitalize on the event did very well. However, for full recovery of species and biodiversity took 30 million years.

So in a major apocalyptic event where 90% of all human and animal life is snuffed out, we might lose large animals like elephants or manatees, but in a few years you'd see trees springing up, rats running around and a limited amount of animals capitalizing on the environment. For new species to appear and all traces of the apocalypse to recover, it would take millions of years.


Here are the issues you need to navigate:

What is the extent of "Recover"?

What is the event?

Answering these drastically changes any answer

If recover means plants and animals exist then it can be anywhere from a few months to a few millenia depending on the disaster. If recover means human habitable the answer could be never. As a specie we depend on certain food sources to survive. If there are no edible plants or animals we cant survive. If the cataclysm wipes out all of them and no new ones evolve to replace them or we dont save any then we are toast.

As for the cataclysm itself: if its a simple meteor that blots out the sun for a year then you could see recovery within a year. If its a big enough meteor to cause a whole plate to flood with lava like what they think is what caused the permian mass extinction, then it could be thousands of years. There are also plenty of man made disasters that could be a killer each with different variables.

--UPDATE-- If the scenario is a nuclear war:

What usually is considered the planet killer is the immense amount of debris and fallout blotting out the sun. Some projections believe this would last for about a year but could be longer depending on how much of the surface is blasted. (if 10% of species survived then it wasn't a total firestorm)

After this period of time you would begin to see recovery as photosynthetic organisms can again begin to live. The radiation is a definite prolonger of this recovery however not an inhibitor. Plenty of plants can survive high levels of radiation. Even chernobyl is covered in plant life now. Animalian life is tricky, some organisms are more resilient than others. This would definitely affect evolution as more resilient creatures would thrive.

  • 4
    $\begingroup$ Usually it is best to ask these questions in a comment and, when the asker has edited their question to answer them, then write your answer. $\endgroup$ Sep 12, 2017 at 18:22
  • $\begingroup$ Except that edit completely changes the question to the point of it being more sensible to make a new question. This at least defines the structure of how to generate that completely different question. $\endgroup$
    – anon
    Sep 12, 2017 at 18:24
  • $\begingroup$ If @SchrodingersStat could just specify what they meant by recover and what the event was (I think the event is some psudo-radiation from a war), that would cover what you wanted, right? $\endgroup$ Sep 12, 2017 at 18:30
  • 3
    $\begingroup$ When the question itself is not clear enough to create a good and informed answer, you should refrain from posting an answer that suggesting to clarify the question. As of now, your answer is very depending on the outcome of the OP's edit. After he edits, you will be forced to rewrite your answer. Hence, it is always be a good idea to post this in comment rather than as answer. $\endgroup$
    – Vylix
    Sep 12, 2017 at 18:31
  • $\begingroup$ By that right, an OP could edit even a concrete question if they wanted and negate everyones answer. $\endgroup$
    – anon
    Sep 12, 2017 at 19:02

Not the answer you're looking for? Browse other questions tagged .