There are tons of small fossils. There’s a 50M year old fish in my son's room that’s about an inch long.

Granted the coral fossil in the living room is many small animals, but it’s not a big object.

Right, I had fossils like that as a kid too, and even had a friend with a ranch in the East Bay that had an exposed fossilized seabed with lots of fossilized shells.

But typically those are extracted from deposits with lots of similar pieces, like the three-foot-long slab of fossilized seabed I had. Professionals might carefully carve something like that up and turn it into lots of small fossils of fish, but they all came from the same chunk of rock.

That's what I meant by "large communities of small things".

So, it's possible that some bunker of physical storage of wealth, with gold bars for instance, will survive a hundred thousand years longer than most of the rest of our civilization. Or, maybe one of our many many landfill deposits will end up with just the right conditions to preserve for a long time the preservable stuff. But I don't think bits of gold jewelry will just be sifted out of the dirt.

Come to think of it, the best bet for evidence of our civilization a million years from now are currently the bits we've left behind on the moon, and Musk's roadster! (The latest calculations have a 6% chance of it crashing back to Earth within the next 3.5 million years -- but it will continue to swing back around every hundred years or so until then.)

I had some interest in this question a decade or two back. What I found from my searches was that it was thought that some human-made ceramics were tough enough to last a million years, pretty much nothing else. How much longer, IDK. Ice sheets will wipe out large structures like the dams and pyramids, but how long will it take to destroy or scatter the remains so that they are not recognizable?

Why though? At least gold is very nonreactive, isn't it?

Sure, but that doesn't make it immune to the passage of time.

Take Mesoamerican gold for instance: only about a thousand years have passed since much of it was made by humans, but it's still considered a rare and valuable archaeological find. (Granted, it was produced in smaller quantities than today.)

Gold is melted by fires; gets scattered and buried by a million years of bioaccumulation; seas move, rivers appear or disappear, continents drift a little, climate changes, glaciers come and go and further destroy evidence.

Google is feeding me a lot of really dumb stuff at the moment when trying to come up with numbers, but I'd guesstimate that most traces of our civilization would be buried a hundred feet below the surface after a million years' time. That makes it pretty difficult to find small things.

Mesoamerican gold was looted and melted down wholesale (along with a lot more silver) by the Spanish. The artefacts were considered unchristian and so were deliberately destroyed, and this has meant that we have barely a fragment of the artistry of the Inca and Aztec.

There is some hope though! It's thought that the melting process was done in Spain so as to allow a reliable royal assay, if so then some of the ships that sank in the trade may have holds full of crushed and squashed precious artefacts.

That’s interesting, can you provide any citations?

Mesoamerican gold is quite common in the gilded church alters of Spain.

It's also extremely soft (for a metal). Fossils survive in part because they are reactive and become harder in certain specific conditions.

It's also very soft

What about steel? Carbon fiber? It’s hard to imagine that fossils survive while a million tons of steel framing for a skyscraper would not.

The process of fossilization by definition replaces organic material with very hard minerals. The fossils we find today are the ones that by chance have fossilized into the hardest minerals under the best conditions for preservation -- buried in the right kind of mud at the time, or by tar more recently, or in ice, or in a bed of limestone.

I can find one not-great reference suggesting that carbon fiber items are expected to be in landfills for on the order of hundreds of years -- not very long in geological scales. Steel rusts and degrades spectacularly easy, and a lot of the steel we've produced is exposed to conditions helping it along.

Neither of these items is susceptible to the process of fossilization; it's that process that helps preserve some things for a long time. The stuff we make has to last that long all by itself, and we're not that good at making things yet.

I'd expect ceramics to survive better than steel. A lot of our archaeological artefacts from former human civilisations are ceramic.