“Natural concentration” is not the right way to look at it because there are higher concentrations that predate the industrial revolution and humans. The all time high (that we know of) is from about 350,000 years ago. This was by all means natural and pre industrial revolution.

"The only known natural concentration empirically compatible with long-term human civilisation"

> empirically compatible with long-term human civilisation

Empirically observed, atmospheric CO2 went from ~320ppm to ~410ppm from 1970 to 2020[0], during which period the human population more than doubled from ~3.7B to ~7.8B and yet deaths caused by climate dropped threefold[1] (not 1/3 the rate; 1/3 in absolute number).

[0]: https://www.climate.gov/news-features/understanding-climate/...

[1]: https://library.wmo.int/doc_num.php?explnum_id=10989

On the scale of human civilisation, 50 years is hardly "long term".

Polonium by that ultra-short-relative-term reckoning is not only harmless as it you still feel fine 10 minutes later, but actually healthsome as you rather feel refreshed by the delicious green tea you just drank in that 5-star hotel bar.

Like the other reply said, there is no natural equilibrium.

We want to return to pre-industrial levels because we’re used to it and we liked it more then.

Really

I like to think of it as scrappy terraforming because we aren't even sure we could handle any of the naturally occurring variation.

Scrappy because, well the planet doesn't quite become uninhabitable and we're starting from the end-game. Science fiction also had me expecting some very cool terraforming infrastructure, not psy-ops to get the serfs to eat bugs.

They are dynamical systems, there is no equilibrium. See also: climate charts for the last few ice age cycles.[1] In the bigger picture we want to modify Earth's climate and definitely do not want to end the current interglacial period, to be fair we've already done that, but returning to a "natural" pre-human climate cycle on the 10,000 year scale is not desirable.

1. https://energyeducation.ca/wiki/images/8/8f/Ice_ages2.gif

Dynamical systems can have equilibrium points —- e.g. an inverted pendulum is stable when hanging straight down. If you deviate too far from an equilibrium point, the system may find another equilibrium that is less desirable for the user. I’m not an expert in climate change, but those things certainly happen for engines, robots, and other systems.