I think it stems from this 2005 conference that basically concluded that at below 400ppm, it would be unlikely that temperatures would rise by more than 2°C above pre-industrial levels and most serious consequences of global warming may be avoided.

"Among the conclusions reached, the most significant was a new assessment of the link between the concentration of greenhouse gases in the atmosphere and the increase in global temperature levels. Some researchers have argued that the most serious consequences of global warming might be avoided if global average temperatures rose by no more than 2 °C (3.6 °F) above pre-industrial levels (1.4 °C above present levels)"

"The conference concluded that, at the level of 550 ppm, it was likely that 2 °C would be exceeded, according to the projections of more recent climate models. Stabilising greenhouse gas concentrations at 450 ppm would only result in a 50% likelihood of limiting global warming to 2 °C, and that it would be necessary to achieve stabilisation below 400 ppm to give a relatively high certainty of not exceeding 2 °C."[1]

[1] https://en.wikipedia.org/wiki/Avoiding_Dangerous_Climate_Cha...

In a way, every point estimate or threshold is "arbitrary" in the sense that it and its consequences are indistinguishable from nearby point estimates/thresholds. So in statistics people usually pick a threshold based on some reasonable risk function and some utility function, and call it a day (but it's important to be clear which functions are being used!). Since there is rarely one right threshold, it is pointless to look for one, but also pointless to complain that the one that was chosen was arbitrary. Unless, of course, it is so far off as to be noticeably less useful in making decisions, and making good decisions is, after all, the whole point of point estimates and thresholds in decision theory.

It's a symbolic number because for as long as we've known about CO2's effects on the climate we have been in the 300s.

It's like the DJIA hitting 10000 or 15000 or whatever. Wooooo we did it people!!!

I wish it was something like a CO₂ meter version of the Y2K bug.

There are interesting side effects, like how as we burn more fossil fuels, we push the frontier where radiocarbon dating is accurate further into the past. But yeah, you're educated, you should know what the effects of increased CO₂ are.

My first idea was https://en.wikipedia.org/wiki/Runaway_climate_change , but that cites a 550 ppm threshold.

I'm more worried about that it increased around 4% on only five years

People in general are comforted by talking about arbitrary thresholds rather than derivatives, probability distributions, etc. It's unfortunate, but thresholding appears to be an effective way to communicate.

carbon is bad, mmmkay?

I like this reference: https://350.org/about/science/