[lsd5you]

Which is problematic.

They are citing it as a cause but have no quantative basis for it what so ever apparently. The effect could be 0, or negative, they've apparently just brought feelings on the situation.

So figure SPM.7 looks pretty serious. It is 60-80 years in the future (probably further at the time of the models) and based on models (albeit on the agreement of presumably independent models). However current shortages are being attributed to climate change. To do this it should be possible to a) show a long term drop in precipitation and b) show climate related causality somehow. This should be easier than accurately predicting the future - but of course is less easy than innaccurately predicting the future.

So what you show I would say reenforces my point. We have predictions based on models (which we can take seriously or not), but then we also have current events which are being attributed to climate change with no quantifiable basis for doing so - despite the fact it should be possible to come up with some kind of proxy for climate induced problems(other than just a shortage happened).

[Certhas]

Actually it's just the opposite.

But literally SPM.7 is the summary of the summary of the summary. There is so so much research published on precipitation, the vast majority of which is careful and precise in giving uncertainties.

It's also actually harder to attribute causality to long term trenda for individual current events than to look at model based predictions for the future. The entire extreme events due to climate change business is incredibly tricky to quantify, even though there are some good heuristic reasons to think it might hold.

[lsd5you]

So you won't conceed that this is problematic that there is according to you presumably relevant data they could cite - even if it doesn't directly apply to yemen in that particular perioud, but they chose not to? I mean what ever happened to 'citation needed'.

I mean uncertainties is a bit of an umbrella term. You should be able to be certain about predictions if you make the error bars big enough. If you still can't come up with a reliable assertion doing this then things are spurious.