Works of Man that Would Last Millions of Years

Underground works like subways require constant maintenance just to keep from flooding and filling up with sediment; there are full time crews that work on the London Tube and New York City Subway systems that do nothing but try to prevent flooding and clean out the system. One of the effects of sea level rise will likely be the abandonment of the subway system under Manhattan Island, parts of which are already below the water table. And the weight of the city above it slowly collapsing as streets cave into the steam tunnels below and buildings fall will serve to compress the tunnels. I’d wager that the system wouldn’t even be accessible at the lower levels within a few decades of neglect, and completely filled or collapsed inside of a century.

Residues of human industrial activity may survive for millions of years; ceramics are essentially fossil-like in composition, so as long as they are buried and not under enormous shearing forces they will survive. The same is true for non-corrosive materials like gold or platinum, and of course high level radioactive waste will be identifiable as from a non-natural origin. However, pretty much any surface structure—even the Giza Pyramids or Angkorian temples—is going to be subject to weathering from sun and rain, the erosive effects of glaciation, and thermal cycling resulting in internal stresses. Structural metals—even the ‘stainless’ steels—will be subject to corrosion, oxidation, and pitting. Although Mt. Rushmore and other exposed earthworks are often cited as being impervious to aging, over geological periods they will break down due to atmospheric and seismological processes. Plastic residues in landfills will turn into an amalgam of complex hydrocarbons as it is compressed by glaciation and deposition, potentially providing a source of rich hydrocarbons and natural gas for future civilizations, but anything based on cellulose or other natural fibers will be broken down into soil.

In space satellites and spacecraft will degrade due to solar and charged particle radiation as well as erosion by interstellar “dust” (mostly ionized particles) over millions of years, and satellites in Low Earth and lower Medium Earth orbit will eventually degrade or break up due to impacts with debris, but the structure of most of the big telecom birds in higher orbits will survive even though the solar panels will be long degraded past any usefulness and the electronics will be completely fried by charged particles. The Pioneer, Voyager, and New Horizons probes will continue on their way, nonfunctional due to the decay of their radioisotope themoelectric generators (RTG) but will otherwise be basically intact. The Apollo, Ranger, Luna, and other spacecraft and components on the Moon will remain in place covered by a slight dusting of regolith deposited by the tenuous ‘atmosphere’ and the occasional meteorite impact, and even the Viking, Pathfinder, MSL, and other probes and landers will remain although significantly eroded by the thin but regular dust storms.

That our most enduring legacy might well be the wastes of once-through nuclear fission power production should be sobering to anyone who considers the significance of human civilization in context.


With enough millions of years all we would have is a weirdly colored layer of rock deep in the ground.

It’s all about luck. But you’ll be amazed by the kinds of traces that can get preserved. There are fossilized raindrop craters from the Archaean. That’s 2.7 BILLION years ago. Of course the current globally- pervasive human presence will be noticeable in a mere few million years. All our junk and the scars of our works would not yet have undergone diagenesis yet - there’s a lot of it in stable, arid continental interiors.

On the scale of millions od years, the “stable, arid continental interiors” may be neither stable nor arid even absent of tectonic shifts. The Quaternary period, starting from about 2.6 Mya, has punctuated warming with periods of major glaciation every 100 ky and minor but still significant periods every 15 to 22 ky. Glaciation will scrape away every surface feature and reform the geology significantly, and the melting as glaciers retreat will carry off sediments in torrents, rerouting or creating new river systems. Fossil evidence may form and paricularly durable materials like ceramics will remain intact but “all our junk and the scars of our works,” will be cleared away by forces that make mountaintop mining seem like a child’s sandbox in comparison.

There is the hypothesis that anthropomorphic global climate change may have ended the Quaternary cycle but not all climatologists agree and some speculate that the intense warming could even presage a massive overcorrection due to interruption of the Gulf Stream, albeit not as quickly as bad science disaster Hollywood movies starimg Dennis Quaid might imagine.


What about the higher radiation levels in items created since the beginning of nuclear testing in the 40s and 50s? I seem to recall reading that testing for such is a reliable means of determining whether the pigments in artworks are from before or after such. Our smart raccoons, God love their little hands, could see that discontinuity and make some sharp judgements from that.

The key to ceramics lasting is that we’ve dumped them everywhere. In 10 million years somewhere that is currently a deposition area will be lifted up and eroding out. Ceramics dumped there will be available for the racoons to find. There’s going to be beer bottles and toilet seats eroding out of hillsides for the next billion (with a b) years.

And, these ceramics will be conspicuously on a geological boundary marking a mass extinction.

Here’s a paper on the subject of detecting a pre-human civilization

“Specifically, Plutonium-244 (half-life 80.8 million years) and Curium-247 (half-life 15 million years) would be detectable for a large fraction of the relevant time period if they were deposited in sufficient quantities, say, as a result of a nuclear weapon exchange. There are no known natural sources of 244 Pu outside of supernovae.”

There would be a sparse layer of long-lived daughter products (chiefly technetium-99) from above-ground testing and high level nuclear waste, but the average distribution will be sufficiently thin it may not be apparent. What would be apparent in the geologic record is a sudden and seemingly inexplicable mercury-dense layer resulting from the inclusion and dispersal of tetraethyl lead in gasoline for in widespread use for over six decades. If these future archeologists are some evolution of terrestrial animals, Procyonids or otherwise, they will wonder why an industrial civilization chose to poison itself in some kind of bizarre global suicide attempt. If they are extraterrestrial, they may just speculate about the presumable protective properties of lead on primate physiology, or that there may have been an external subterfuge to undermine civilization. It will surely not occur to the that a civilization would deliberately do this to itself just so some small minority could benefit financially.


Well jeez Louise, we’ve got to leave them somethin’ or they’ll never know we were here!

The pyramids are missing significant amounts of their original structure now, after mere millennia. By a million years out, they’ll just be more sand in the deserts.

Eiffel 65 “Blue”

ducks the flying rotten cabbage

The outer layer- which was looted.

They’re still basically hunks of stone in a sandblaster. No way they make a million in anything like recognizable form.

What about the silicon strata of microchips? There pretty much ubiquitous and if even a piece of one was found an advanced civilization should recognize it.

Left in a climatically controlled area/deep beneath the earth?

A Big Mac. Seriously.

The Namib has been an arid desert for at least 55 Ma. See also the Atacama. And those are on the coast. We have newer deserts now on stable cratons - who knows how those are going to pan out in the Anthropogene.

My guess is anything than can fossilize, or be preserved similarly (for non-living items that is.)

“Ice ages” are sufficiently sensitive to the precise alignment of continents and connectivity of ocean basins that I don’t think they can be predicted over a scale of tens of millions of years. It is hypothesized that the Pleistocene glaciations really got going due to changes in ocean circulation due to the closure of the Isthmus of Panama about 3 million years ago, although narrowing of the connections between the Indian and Pacific oceans by movements of Australia have also been implicated. At any rate, recent continental glaciations have not reached past mid latitudes. Unless we are talking about a “snowball earth,” glaciations aren’t going to clear away surface features in tropical or subtropical regions.

Now I’ll grant that even in the mid-latitude arid belts the remains of buildings or human-made constructions might not not be visible on the surface due to erosion, sedimentation, or other factors after a few million years. On the other hand, remains of cities could be buried (and current rising sea levels makes this likely), protected from erosion, and then exposed again through uplift. We have plenty of examples of fossilized reefs hundreds of millions of years old that are now exposed on the surface, and they are made largely of soluble material like limestone.

With regard to regional climates in a future Earth, it is worth mentioning that even 100 million years in the futuremost continental land masses except for Australia won’t have moved far from their present latitudinal locations. Arid areas are determined partly by atmospheric circulation and partly by how remote an area is from the oceans, so some areas like Central Asia are likely to remain arid over even this time period. The next supercontinent, which has been dubbed “Panagea Proxima,” may not coalesce for another 250 million years.