Hacker News Reader

30 years ago (1988), Maldives was predicted to be underwater by now. These kind of extreme "predictions" which don't pan out makes regular folks doubt the science. There need to be percentages of the event actually happening assigned against any forecast.

View on HN

Bingo. This is a problem every time scientists run to the media to explain the latest new bad discovery.

This. The covid predictions have been utterly wrong so far based on Imperial College's models.

Can I trust climate change models more and why should I?

yes you can.

the reason is that epidemics grow and shrink more or less exponentially - if you mispredict a particular point in time, the error from that point onwards also grows exponentially. temperature relationship between CO2 and other GHGs are approximately linear. global warming predictions made 40 years ago about average temperature rise are basically spot on.

I have seen plenty of predictions from the 80s or 90s that didn't come true. So it would seem that there are multiple models. How do I find out what ones have been correct in their long term predictions?

IPCCs models from 90s that predicted the average temp rise in 2010's are spot on, you can look into them maybe?

That's a start, I will do.

Reply on news.ycombinator.com