Humanity just disappears tomorrow, let's say it is like the Thanos-snap but it kills everyone. How long until the last unambiguous evidence of our existence disappears? What was this last relic? The faint spectres of electromagnetic signals which will travel space forever, odd isotopes ratios and material compositions and slightly odd geological patters are considered ambiguous. A Voyager probe which still looks like something designed by intelligence and not like a very metal-rich asteroid is unambiguous.

Assume that another civilisation's exploration mission with near-future tech and the capability to examine the solar system for 100 years comes looking. They got the time and huge resources on hand and will study the major objects until they know about them as much as we currently know about Earth.

  • 1
    $\begingroup$ Comments are not for extended discussion; this conversation has been moved to chat. $\endgroup$ – L.Dutch - Reinstate Monica Jul 20 '19 at 19:52
  • 2
    $\begingroup$ The last unambiguous evidence of humanity will be an Ericsson GA628, which will last until the heat death of the universe. $\endgroup$ – Whelkaholism Jul 22 '19 at 11:05
  • $\begingroup$ I've read that fallout from nuclear explosions will allow one to pinpoint not only that, but also when there was intelligent life on earth for a very long time. I don't know if that would be the last artifact, or how accurate it is. $\endgroup$ – Jasper Jul 22 '19 at 11:54
  • 3
    $\begingroup$ @Jasper Radiation is the gift that keeps on giving. $\endgroup$ – VLAZ Jul 22 '19 at 12:13
  • 1
    $\begingroup$ @Alma I believe it's pretty much agreed in the scientific community that the sun won't go supernova but will go red giant and then later white dwarf. I don't think much humans have left will be able to survive the red giant stage, though, unless we have started leaving things further out in the solar system at the least. $\endgroup$ – Jasper Jul 22 '19 at 18:07

12 Answers 12



With orbital multispectral imaging available there is no way that any remains of our cities will be missed. Some of those remains won't last long, geologically speaking, but others will last through multiple cycles of super continent formation and break up. Although these traces will be small relative to the scale of modern construction, they will present unambiguous grids of mathematically straight lines many miles across where old road surfaces interrupt soil formations and buried walls and rubble piles disrupt plant growth. The cities that last the longest are going to be the ones that are in the middle of the continental building blocks know as cratons; Alice Springs in Australia and Hyderabad in India are prime suspects for leaving traces the longest as are Moscow, Riyadh and Brasilia. Those are probably going to be detectable, at first scan, with current/near-future technology while the world lasts. With a century to scan and check data, I'd consider it a near certainty. Unless Earth gets hit by another heavy bombardment: that would almost certainly cover up all evidence of human habitation.

  • 2
    $\begingroup$ This is basically what I was thinking. Aerial and Satellite imaging tech we have right now is finding evidence of structures humans built thousands of years ago with stone tools. The marks we're leaving behind right now in concrete and steel aren't going away even on geological time scales. livescience.com/… $\endgroup$ – Morris The Cat Jul 19 '19 at 16:51
  • 1
    $\begingroup$ @MorrisTheCat Weirdly enough modern reinforced concrete will go the way of the dinosaurs faster than the old stone and earth constructs of the neolithic. $\endgroup$ – Ash Jul 19 '19 at 16:56
  • 2
    $\begingroup$ Even harder to miss humans if we get structures established on Mars: less tectonic activity, less weather decay, less bacterial decay. Actual structures might survive there, and might well mention Earth as our origin point. $\endgroup$ – SRM Jul 20 '19 at 13:46
  • 3
    $\begingroup$ Note this will disappear within a fer million years, just due to burial and erosion. It is ambiguous evidence at that point. $\endgroup$ – John Jul 20 '19 at 15:15
  • 2
    $\begingroup$ You need to find cities, where (once people are gone) material is starting to accumulate on top of the human city. If there's just 1 mm of erosion per millenium, even tunnels and pipes 100 meters deep will be gone in mere 100 million years. I'm sure there are such cities, but climates and continental movement changes chaotically, so will any remain this way in geological time scales? $\endgroup$ – hyde Jul 22 '19 at 11:05

If enough time goes by, there would fewer and fewer signs on the surface of the Earth for someone looking for past intelligent life.

However, as soon as someone wants to build any infrastructure or industry on this planet, they'd notice the distribution of metal is very strange.

All the easy-to-mine metal is gone!

enter image description here

Even long after every mine on Earth caves-in, they'll find the less profitable metal with large sections of the metal veins missing, as if someone came by and took all the good stuff before they got there.

Once they notice how odd that is, they'll start looking around for signs that some Earth-unique process could have moved it somewhere, and even start to wonder if someone else was here before them. Looking for metal in un-natural places will quickly reveal engineered alloys that do not exist naturally.

Once they see the alloys, they'll start mapping the discoveries and they'll look for patterns for where the high-value metal is located. They'll find whatever is left of our cities and garbage dumps. The smoking gun will be the screws and nails that can be found everywhere we lived, long after the buildings have been lost to time. We'll probably be know as the "screw and nail users". The proximity to water will become obvious. They'll start looking for us along the ancient coastlines and old dry rivers, on the shores of what could have been lakes or valleys, and anywhere that matches our pattern.

Eventually they'll look through the fossil record of formerly-inhabited places and narrow us down to being either humans, dogs, or cats though a process of elimination.

  • 8
    $\begingroup$ This is a very good point. Open pit mines, deep mantle mines, coal mines - slag piles.... The Centralia Coal Seam fire might still be burning and vegetation has a lower erasing-effect the deeper you go. It's not just the metal distribution, it's the geological oddities (did glaciation cause this? Probably not...) and advanced visitors would certainly know the difference. The evidence of mining may never go away. $\endgroup$ – JBH Jul 20 '19 at 13:42
  • 9
    $\begingroup$ The opposable thumbs will quickly give us the nod over dogs/cats. That, and every relic built to our scale and height. $\endgroup$ – SRM Jul 20 '19 at 13:55
  • 6
    $\begingroup$ @John: No depletion after a few million years? No, not even close. Most coal mined today was laid down during the Carboniferous, three hundred million years ago. Most likely it will never be replaced. Metal deposits were created by geological processes as old as the rocks they sit in. Our deep gold mines plum rocks that are over 2.5 billion years old. $\endgroup$ – Jack Aidley Jul 21 '19 at 8:41
  • 3
    $\begingroup$ "All the easy-to-mine metal is gone" how would you know something that isn't there now was there previously? Maybe there never was any "easy-to-mine" metal? $\endgroup$ – freedomn-m Jul 22 '19 at 14:12
  • 2
    $\begingroup$ Even moreso than this, any future archaeologists would notice the isotopic ratio of the planet has been very heavily skewed due to nuclear power. Even now, two billion and some change years later, we can detect the Oklo nuclear reactor, and that was natural. They would notice that some deposits of uranium are depleted in isotopes, while different isotopes of other metals such as cesium, strontium, barium, and lead are enriched in certain areas. Even without being able to find a single artifact, they would know something strange happened. $\endgroup$ – stix Jul 22 '19 at 16:51

probably longer than the earth has.

Fossilization is a thing, and we have set up many things in perfects places to be fossilized. We bury things in salt mines, seal things in glass, bury massive amounts of garbage in anoxic conditions, etc. On top of that we have built things that will leave traces for billions of years, chernobyl will stick out like a sore thumb, much like how Oklo did when it was first discovered, but the composition of materials will make it very obvious it is not natural (corium). Oddly some of our oldest inventions will be the most obvious stone tools and fire. Humans have created a lot of fires but often in small isolated stone lined pits, these will be very obvious markers in sediment for billions of years, even more so when combined with obsidian fragments and cut bone. Even things like buildings can "fossilize" that is get buried and preserved, exposed building will not last long but someplace like pompeii or missile silos built depositional environments wi last as long as the rock in those areas do.

Then you have the direct evidence from out own bones, humans are everywhere some of our bones will fossilize and some of those will will have dramatic evidence of technology. Not just the evidence from out anatomy but things like false teeth, polymer and ceramic implants, glass eyes, plastic buttons, glass lenses, etc. One fossilized humans skull with ceramic dental implants will be unambiguous proof of a technological species. So not only will they know a technological species existed, they will have a decent idea of what it looked like.

The only way to destroy all this evidence will be to destroy earths surface geology which means liquefying the planet which is unlikely to occur before the sun explodes, or even for a while after that.

  • $\begingroup$ I was sort of thinking along the same lines, though in terms of the fossil record of the Holocene extinction event - which seems to me to be harder to miss than a mere million years or so of hominid and human-created fossils. Which would look quite unusual compared to past extinction events, because of it coinciding with a sudden worldwide spread of many plant and animal species. $\endgroup$ – Daniel Schepler Jul 20 '19 at 0:03
  • $\begingroup$ @DanielSchepler I tried to focus on unambiguous traces, we have certainly left a lot more marks of global geology than just what I discussed, but many of those could be confused for natural events, especially after so much time. Stuff that would make any scientist worth his salt go "damn there was a technological species here!" I mean if we found a fossil dinosaur with a hip implant it would be hte single greatest discovery in human history. $\endgroup$ – John Jul 20 '19 at 3:44
  • $\begingroup$ I read somewhere that (billions of) spark plugs (the ceramic insulating part) would survive forever. I don't know why these would last longer than glass or pottery. $\endgroup$ – Keith McClary Jul 20 '19 at 3:54
  • $\begingroup$ @KeithMcClary The only thing that worries me with spark plug ceramics is them being mistaken for a biological product, like shells. They are small and the shape is not obviously technological. $\endgroup$ – John Jul 20 '19 at 15:18
  • $\begingroup$ The classic artifact is the Coca Cola (TM) bottle - the old glass ones. Being glass, they will last geologic ages without disappearing, and they were designed specifically to allow identification from a fragment. I've heard archaeologists humorously projecting future archaeologists as writing papers about the relative differences between CocaCola Man and Budweiser Man. $\endgroup$ – WhatRoughBeast Jul 20 '19 at 23:34

The Earth After Us by Jan Zalasiewicz is a book-length treatment of this very question. My main takeaways were not to underestimate the power of erosion, but that some evidence would last a very long time.

The lifetime of Mt. Rushmore, as I recall, was measured in the millions of years at most, not hundreds of millions. The book also made the argument that city layouts would not last and would be completely eroded away if not buried. But as I read up on cratons I see Ash's point.

On the other hand, much will be buried and will become part of the fossil record. Dinosaurs were around for hundreds of millions of years, but most of their bones decayed before fossilization. We are leaving behind a huge amount of buried ceramics and concrete.

And the Apollo sites will be on the moon probably for almost as long as the moon survives. Correction: Apparently that's wrong. According to this article in Space.com, "From past studies of moon rocks collected by astronauts during the Apollo missions, researchers have learned that the rocks erode [from micrometeoroid impacts] at a rate of about 0.04 inches every 1 million years.", That's four inches of erosion in a hundred million years, and the landing stage is not a solid hunk of rock. Oh, well.


Mega Construction

We build, not just cities, There are dams huge chunks of smooth cement laced through with even grids of steel. The edges of the Hoover Dam and Three Gorges Dam and Aswan Dam will be there long after the rivers run dry.


We dig. There is an open pit mine 3 miles long, 2 miles wide and half a mile deep. It's not the only one. There are similar mines on every continent except Antarctica. The sharp edges of the mine will eventually erode but the shape of the land has been indelibly changed. The spacing of the mountains and ridges will be inconsistent.


Satellites in Geostationary orbit are up there for millions of years. If aliens come along they will see a large number of items floating suspiciously close to a synchronous orbit and careful collection should find something relatively intact.

Space Probes

Most of the above should be detectable until the sun starts to expand and swallows the earth. There are currently 5 space probes which have achieved escape velocity from the solar system. They should be observable pretty much forever.

  • 1
    $\begingroup$ Hoover Dam won't last: the Colorado River (without dams) is known for violent flooding, so it'll erode quickly once it's breached. The Nile's floods aren't as violent, but they're persistent, and Aswan is an earthfill dam: it won't last much longer. Of the three dams you list, Three Gorges is the most likely to survive, being a gravity dam in a relatively broad valley. Even it's only going to be good for a few thousand years of shifting river flows. $\endgroup$ – Mark Jul 19 '19 at 19:48
  • 2
    $\begingroup$ @Mark: The dams themselves won't last, but the evidence of their former presence (deep, relatively-uniform sediment deposits upstream of the former dam location and extending well beyond the edges of the river, with only thin or no corresponding deposits from the same time periods immediately downstream; riverbed deposits and erosion patterns showing extensive disruption to, or even complete absence of, the normal seasonal flood patterns downstream of the dam; possibly the remains of the spillway(s) and/or diversion channel(s) [filled with sediment(ary rock), but still existing... (part 1) $\endgroup$ – Sean Jul 20 '19 at 23:27
  • 1
    $\begingroup$ ...as disruptions in the preexisting native strata], depending on how the dam was designed and built; sedimentary and erosional evidence of single-occurrence, cataclysmic flooding at the time of dam collapse, followed by the resumption of normal [by pre-dam-construction standards] sedimentation and erosion patterns; evidence of a sharp discontinuity between flow volume upstream and downstream of the dam during the dam's existence [becoming more pronounced as the dam silts up - progressively blocking more and more of its outlet channels - in the leadup to final collapse]; possible... (part 2) $\endgroup$ – Sean Jul 20 '19 at 23:32
  • 1
    $\begingroup$ ...chunks [potentially quite large] of remaining dam material embedded in the sediments of the dam-collapse sequence, depending on the dam's construction; etc., etc., etc.) will, potentially for a very long time indeed. Agreed, though, that that probably isn't what MongoTheGeek was thinking of when they wrote this post! (part 3) $\endgroup$ – Sean Jul 20 '19 at 23:34
  • 1
    $\begingroup$ I envisioned large chunks of the sides embedded in the walls of the canyon still attached after the dam and it's sediment bed eroded beneath it. $\endgroup$ – MongoTheGeek Jul 20 '19 at 23:52

When The Earth Dies

Naturally occurring zircons have been dated to 4.4 billion years ago, or basically right after the earth cooled enough to form a solid surface: https://www.livescience.com/43584-earth-oldest-rock-jack-hills-zircon.html. Therefore, it is possible for certain tough minerals to survive for basically as long as the planet does.

I would imagine that a fair number of cut and manufactured gemstones would have similar survivability, although the difficulty of locating them after millions of years will likely go up exponentially. However, a single cut diamond with a laser-engraved serial number would be pretty definitive evidence of technological capability, even if the number were partly damaged.

Oldest Sediments

Gnudiff claims that the geologic record doesn't extend past 2.6 mya, but we have recovered rocks dated to more than 150 mya: https://earthscience.stackexchange.com/questions/3115/what-is-the-oldest-sediment. Granted, if space invaders visit earth 100 my from now looking for us, and we have been gone the whole time, they will need to have a very fine-toothed comb to discover any gemstones, but it is at least theoretically possible.

  • $\begingroup$ Note that that question on Earth Science SE is referring to the oldest sediment, not the oldest sedimentary rock. There is a huge difference between the two. $\endgroup$ – Gimelist Jul 21 '19 at 13:32
  • $\begingroup$ We have recovered fossils of animals over half a billion years old. I am pretty sure those were buried in sediments. $\endgroup$ – hyde Jul 22 '19 at 11:25

The Atlantic describes exactly this question in an interview with NASA's Gavin Schmidt.

The answer is: "When it comes to direct evidence of an industrial civilization—things like cities, factories, and roads—the geologic record doesn’t go back past what’s called the Quaternary period

2.6 million years ago.

For example, the oldest large-scale stretch of ancient surface lies in the Negev Desert. It’s “just” 1.8 million years old—older surfaces are mostly visible in cross section via something like a cliff face or rock cuts. Go back much farther than the Quaternary and everything has been turned over and crushed to dust."

So basically around 0.04% of current Earth's age.

See also link to the study referenced in the article.

Now, if you are talking about indirect evidence, there is some that can be discerned for much longer (also described in the article). However, it seems to me to fail your criteria of "unambiguous".

Actually, depending on your requirements for unambiguous, it could be on an order of magnitude less, if you wish your aliens to actually learn a lot about our current existence, not just that something existed.

So, maybe 200K years for something recognisably current human civilisation.

  • $\begingroup$ There are several fossil beds dating over half a billion years (from Ediacaran and Cambrian) with quite fine details of animal remains preserved... in 500 000 000 years there will be simlarly fossilized human skeletons, construction bricks, a lot of cars, insane amount of various plastic item shaped fossils... $\endgroup$ – hyde Jul 22 '19 at 11:23
  • $\begingroup$ @hyde I guess the key point here is "several". Ie the fossils have only a sample of small minority of life on Earth. They form only under specific conditions and bones otherwise decompose in under 100 years, unless in desert conditions where they can survive several K. It is surely possible to contrive a story where the remains of human civilization do survive for the whole remaining history of Earth, as one answer suggests. However, is it likely? A scientific paper tells us, not really. You can argue with the paper, however, it might be useful to see what have they considered first. $\endgroup$ – Gnudiff Jul 22 '19 at 12:06
  • $\begingroup$ @hyde Additionally, as I noted in my answer, it depends on what OP considers unambiguous. Finding half a dozen fossils over Earth surely can be said to be evidence of humankind, however, based on OPs examples I assumed he was talking more about preserved things of our current civilization which could be used to get some idea of how we lived. $\endgroup$ – Gnudiff Jul 22 '19 at 12:08
  • $\begingroup$ But these fossil sites cover millions, even hundredes of millions of year. Also, consider how we have discovered iridium layer in sediments caused by meteor impact at KT boundary (dinosaur extinction). These aliens spending 100 years researching the Earth would find global "plastic layer" or carbon compound layer with unique chemical and isotopic (fossil fuels) signature, and after discovering that they'd know to look for sediments of this age to find out WTF that is about. And then they'd start finding more than just chemical signature of plastic, when they'd know where to look. $\endgroup$ – hyde Jul 22 '19 at 22:17
  • $\begingroup$ @hyde True, but OP's stated problem is all of humanity goes extinct in one day, so whether that day is preserved across millions of years, or some other days, becomes rather an important question. Of course, you are right about plastic layer and other things. Whether my answer stands rather depends on what OP considers sufficient unambiguous evidence of our civilization. Plastic layer will probably exist, but what can you deduce from it? We find dinosaur fossils, and we have deduced a whole lot from them, but how sure can we be about what they lived like? $\endgroup$ – Gnudiff Jul 23 '19 at 16:41

Enriched Uranium

Uranium 235 has a half-life of 703,800,000 years. The concentration of U-235 relative to U-238 is fairly low (0.72%) in nature. Humanity has managed to increase that quite a bit. Perhaps a bit more than we should. In any case, even if the surrounding concrete, steel and other materials of research & power nuclear reactors, nuclear submarines, nuclear aircraft carriers and other nuclear devices (not so much the nuclear weapons - if they use plutonium then the half-life is much shorter at a mere 24,100 years) have disintegrated due to various natural processes, the enriched uranium will still be detectable as a non-natural object, for a very long time, easily billions of years.

  • $\begingroup$ Pu-244 has a half-live of 80my. $\endgroup$ – Martin Schröder Jul 26 '19 at 21:31

New Scientist produced an article on this on 11 October 2006, where they suggested that buildings would be overgrown fairly quickly (decades in many cases) but that ruins would leave evidence for thousands of years if searched for. Human effects on climate would continue for at least a century, although these probably don't count as unambiguous according to OP. Large monocultures such as grain fields or single species forests would take centuries to become diverse.

However, they also suggest that even 100,000 years from now, there would be signs of humans, as described in this quote from the article.

"Yet if the aliens had good enough scientific tools they could still find a few hints of our presence. For a start, the fossil record would show a mass extinction centred on the present day, including the sudden disappearance of large mammals across North America at the end of the last ice age. A little digging might also turn up intriguing signs of a long-lost intelligent civilisation, such as dense concentrations of skeletons of a large bipedal ape, clearly deliberately buried, some with gold teeth or grave goods such as jewellery.

And if the visitors chanced across one of today’s landfills, they might still find fragments of glass and plastic – and maybe even paper – to bear witness to our presence."

However, even that is not the limit. By some measurements satellites at a height of 10,000km could still be around 10 million years from now. They won't work, but they will be obviously artificial.


First category: Artifact of man-kind on Earth:

Due to fossilisation they will probably survive until the Earth is swallowed by the than red-giant sun. The last unique relics may be teeth with gold inlays, clearly showing that something special was going on with our species. Expected timeframe: 4.5 billion years from now.

Second category: Artifacts in space:

Unfortunately, almost all artifacts in space will also be swallowed by the red giant sun at about the same time. There are a very few exceptions: The Voyager probes leaving the solar system, and a few probes sent to the outer part of the solar system. However, will they be attributed to some species once populating planet Earth?



The last thing to go will probably actually be Mount Rushmore it's solid, meta-static (meaning in pressure equilibrium and not subject to foliation erosion), granite in a geologically stable region with reasonably stable continental weather patterns. I've never heard an estimate for how long it will take to be unrecognisable but I have seen it estimated that it definitely will still be recognisable in 1.42 Billion years time. This is long after Voyager and the Moonlanders are expected to have been destroyed by micro-meteor impacts, neither of those, nor Neil Armstrong's foot prints, are expected to last until the next super continent formation in 100-200 million years.

  • 14
    $\begingroup$ I feel like the supervolcano under Yellowstone might have something to say about that though.. $\endgroup$ – Morris The Cat Jul 19 '19 at 14:19
  • 12
    $\begingroup$ A billion years is time enough for a mountain range the height of the Himalayas to be eroded to the rolling hills of the Alleghenies, twice over. No rock currently visible on the surface of the Earth can be counted on not to be lost in that length of time. The ancient rocks that have survived were not at the surface the whole time, and burying and unburying will also destroy. $\endgroup$ – Mark Olson Jul 19 '19 at 14:22
  • 3
    $\begingroup$ I am pretty certain that out the vast quantity of faceted gems distributed worldwide some will outlive that specific piece of high-relief statuary. $\endgroup$ – AlexP Jul 19 '19 at 14:29
  • 3
    $\begingroup$ Granite is tough, but even in optimal conditions it will be eroded in a few millennia. South Dakota has enough rain and temperature swings to exacerbate erosion. After the next ice age eventually comes, there is zero chance that the human carvings there would survive. My bet is that Pyramids of Giza would outlive Mount Rushmore (as a recognizable artificial feature). $\endgroup$ – Alexander Jul 19 '19 at 16:25
  • 2
    $\begingroup$ 30°C? Very easy for southeast-facing rocks in continental climate. The mineral does not need to be permeable - water enters through minor cracks during the day and freezes overnight. And I would stress again that if (or rather when) the glacier comes, the game is over for Mt. Rushmore. $\endgroup$ – Alexander Jul 19 '19 at 18:06

I'd guess that digging will always indicate a really strange segment of sediment from this period. Probably remains of plastic and other unnatural chemicals with a really long life. Also the fact that we pulled up so much carbon and laid it down over such a short period should be something to look for.

I don't think it will be destroyed until the earth is consumed by a star.

Even then--information can't be destroyed. The state of any system can be wound backwards given complete information about the state. Of course I can't even imagine something that could wind the universe backwards, but theoretically...


Not the answer you're looking for? Browse other questions tagged or ask your own question.