2011-05-12 02:06:10Carbon Brief post - Is CRU the principal source of climate projections?
Robin2
Robin Webster
robin@carbonbrief...
2.25.45.111

hi, this is Robin from the new UK-based website, Carbon Brief. We really admire what you're doing at SkS and have been talking to John about how we can colloborate a little bit. John kindly suggested that we can cross-post some material, which we are now intending to do over the next few weeks. As the first of these, we're hoping to shortly cross-post this with SkS - Is CRU the principal source of climate projections? 

Perhaps not the most exciting of posts to start with (!) - but any feedback on this post over the next few days (probably cross-posting Monday next week) would be really appreciated. Thanks!

2011-05-12 02:53:37
MarkR
Mark Richardson
m.t.richardson2@gmail...
134.225.187.197

It reads fine, but what about some pictures?

Can you use googlemaps or something to make take a map of the Earth and put on where the groups are? And maybe a graph from one of their climate model outputs or their temperature record on each one? Might be too much work, but it could have a lot more impact than text alone. :)

 

 

 

Also, is this:

"As has already been profiled both here and on our site at Carbon Brief, there are three principal surface temperature datasets, of which the CRU/Met Office hold just one."

true? I count 4: NASA GISTemp, NCDC, HadCRUT3 and JMO.

2011-05-12 10:57:29Fixed links
John Cook

john@skepticalscience...
124.187.6.188

Many of the links only had one slash in the http (eg - http:/www instead of http://www) which causes the links to go funky. So I fixed that. I also indented the quotes for some visual structure.

Robin, I've upgraded your account to "Full Author". This means you can publish this blog post on SkS - when you edit it in Author Admin, set the status to Published. This way, you can go live with this post at the same time you post it on CB (plus I'll be travelling at the time and won't be able to do it myself).

2011-05-12 20:10:43
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.1.168

Looks good to me, but still plenty of broken links. On the topic of the multiple model runs, anyone know if they've made any progress on why the average of the various models, fits reality better than any individual model?.

2011-05-13 00:27:04
MarkR
Mark Richardson
m.t.richardson2@gmail...
192.171.166.144

Rob: isn't that expected to happen anyway?

Assuming each model selects a set of parameters from parameter space with equal probability, and that the real world has also a set of these parameters, <i>and</i> that the cumulative effect of these parameters leads to a Gaussian distribution of climate variable vs # of parameter choices, then an ensemble is more likely to resemble the real world than any individual model!

2011-05-13 05:45:45
Rob Painting
Rob
paintingskeri@vodafone.co...
118.93.7.155

Mark - don't know, but it was watching interviews with climate modellers that caught my attention. I think it was Gavin Schmidt, that was discussing this. He seemed at a loss to explain it. Mind you, this is a few years back now.

2011-05-13 06:36:44
dana1981
Dana Nuccitelli
dana1981@yahoo...
69.230.97.203
Good post. A few suggestions - I'd combine the second paragraph with the first. I agree with Mark on adding a few figures, and would suggest the SRES CO2 concentration projections which we've got here, and a comparison of the temperature data sets which we've got here. I'd also suggest mentioning AR4 was published in 2007, and tying the conclusion back to the point that CRU isn't the pinciple source of climate projections.
2011-05-14 01:30:36
Robin2
Robin Webster
robin@carbonbrief...
88.211.15.182

thanks! Comments useful and much appreciated. We're just doing some final edits in response.

On the comment about the number of datasets, I was trying to be a bit careful as we talked about four (including the JMO) and the SkS post I linked to talked about the three surface datasets (NASA, NCDC and HadCRU). So I just said three surface datasets.

2011-05-14 02:41:09
Robin2
Robin Webster
robin@carbonbrief...
88.211.15.182

We've now put an updated version up, in case anyone wants to look at it over the weekend before we publish.

2011-05-15 07:07:29A concern
John Hartz
John Hartz
john.hartz@hotmail...
98.122.68.19

I have a concern about the following statement:

"The climate observations provided by these datasets can be used to test climate models, which are able to reproduce observations of past and current climate reasonably well - indicating that their large-scale projections of future climate change are also likely to be reasonably accurate."

If historical and current datasets used to calibrate the models. there darn well better be a "good fit" when the models are used to "hindcast." I therefore question whether this statement is defensible.

2011-05-16 20:39:31
Robin2
Robin Webster
robin@carbonbrief...
88.211.15.182

Thanks. We didn't quite mean that as the sentence didn't refer to calibration of models, but could see your point in terms of how it read. We have taken it out as the article isn't directly about the reliability of climate models.

2011-05-17 01:32:35
dana1981
Dana Nuccitelli
dana1981@yahoo...
64.129.227.4

Looks like you published this on Carbon Brief already.  Is it ready for publication here too?

whoops nevermind - I didn't see that it already was!