That's the optimistic timeline :)
I've read somewhere that up to half of Antarctic sea ice could easily disappear in hours or days, given right conditions. Greenland is not looking good either.
You use a wooden chair for reaching up in high places. Imagine you're a short person and utterly dependent on this chair. It works for 20 year, and it helps you reach what you need in all that time. You never notice much except for some familiar creaking. The chair works every time, and you just take it for granted as a handy tool to have around.
But when the wood in that chair snaps, it happens all at once. Even if just a single part of that chair momentarily fails. Gravity, leverage, falling backwards with no way to protect your head. And it happens so fast, there's no time to think about it while it happens.
Luckily your husband put a cushion on the floor exactly where you'd fall with your neck, because he anticipated this happening. He just didn't know when it'll happen.
You wake up a bit dazed, and a bit wiser to how the world works. How fragile and complex systems can fail all at once and too fast.
You fall, and the floor gives way. You plummet through the floor, down to the basement, and sever an artery on the way down.
Crop failures, famines, civilization collapse, wars, violence, slavery ...
People have been talking about that for a while, we're playing god right now but we have no idea what's coming. It's the same thing every time we try something with nature, what seems easy and simple turns out to have massive side effects we couldn't imagine because every little thing is inter dependant
This is what worries me: climate change leading to entirely new regimes that we don't understand and thus can't model or predict with any accuracy. You just can't build a seawall for a sealevel rise of between 50 and 300 cm somewhere in about 20 to 50 years. Or for heatwaves of maximum 35 to 45 degrees, precipitation of max 20-100mm/hour, etc. Climate adaptation will be even more challenging if we don't know what we are adapting to