Hacker News Reader

Comment by trevelyan | Hacker News Reader

Predictions of global warming go back to at least the early 1990s and clearly precede findings like this:

http://www.sciencedaily.com/releases/2007/12/071213101419.ht...

Yes, but I haven't seen specific predictions of local chills, only post factum explanations. Consider an excerpt from the article you posted: "2007 started with record breaking temperature anomalies throughout the world. In parts of Europe, winter and spring ranked amongst the warmest ever recorded, with anomalies of more than 4°C above the long-term monthly averages for January and April." They're using a short term local warming as evidence for global warming here, however the original article clearly states you can't do that (and explains the chill away in this manner).

"Yes, but I haven't seen specific predictions of local chills, only post factum explanations."

This is actually a textbook difference between Climate and Weather. We're better at predicting the former than latter.

Are we? Is there a study that compares a sophisticated climate prediction model with a trivial model (one that simply adds the delta from the previous two years, for example) and demonstrates that in comparing each result to the real temperature the former gives a statistically significantly better result than the latter? That's the only way to know if we're any good at predicting climate. If you can point me to a study on this (or some review study) I'd really love to take a look.

> This is actually a textbook difference between Climate and Weather. We're better at predicting the former than latter.

According to the text book, warm days are climate while cold ones are weather.

Reply on news.ycombinator.com