|
Post by thermostat on Aug 2, 2011 4:05:17 GMT
Sigurdur, You are mistaken. But please, feel free to provide scientific references. I would be quite amazed to see substantive science that supports your confused opinions. I am sure that you would be amazed if you had actually read the literature as you claim. No....on this one the impitus is on you to prove me wrong. And please, keep us informed of your progess by posting links to the literature. Sigurdur, I keep waiting for you to provide a useful scientific citation but I appreciate this is not what you do. Sigurdur posting links to the literature?
|
|
|
Post by thermostat on Aug 2, 2011 4:28:47 GMT
Sigurdur, You are mistaken. But please, feel free to provide scientific references. I would be quite amazed to see substantive science that supports your confused opinions. I am sure that you would be amazed if you had actually read the literature as you claim. No....on this one the impitus is on you to prove me wrong. And please, keep us informed of your progess by posting links to the literature. Sigurdur, 'Reading the literature' is not the daunting task you seem to imply, but rather an every day activity among us nerd scientists. We do it for fun; I'm not kidding. Forum members appear to be quite confused about all of this, ie how scientists think and how they behave.
|
|
|
Post by steve on Aug 2, 2011 11:06:32 GMT
icefisher
I agree that one of Spencer's main aims is to criticise *other peoples* observation-based assessments of climate sensitivity. The problems with this are two-fold:
1) you have to read past a lot of fluff (the press release and press hoo-ha that did not represent the paper accurately, whatever Roy likes to pretend) before you get to this point. 2) His method of estimating climate sensitivity is so obviously faulty it doesn't really address other people's methods.
It's not relevant to say that he has addressed issues in his blog, because they should have been addressed in the paper. Saying that "the reviewers didn't think they were important" is not good enough if people like me can identify some of the issues.
|
|
|
Post by steve on Aug 2, 2011 11:20:06 GMT
sigurdur
The IPCC figure (2-4.5C by 2095) is based on the range of model responses to the emissions scenarios.
The range of model responses is validated by the model validation process (comparing pre-agreed model diagnostics with observations) and assessments of climate sensitivity. Assessments of climate sensitivity are derived from paleo data and from assessments of recent observations. These observations put a low end sensitivity of 1C.
I don't really want to have an argument about whether these processes are good enough, but they cannot be dismissed in the way that you have done. I can see that you, Spencer and Christy are similar in that you all like your simple home-spun tales for how the climate should work and how things should be done, but it doesn't really reflect either the reality of the climate or the reality of how climate research is done.
That's an easy one. Just because A can happen without B does not mean that A cannot also be caused by B.
|
|
|
Post by magellan on Aug 2, 2011 16:26:43 GMT
sigurdur The IPCC figure (2-4.5C by 2095) is based on the range of model responses to the emissions scenarios. The range of model responses is validated by the model validation process (comparing pre-agreed model diagnostics with observations) and assessments of climate sensitivity. Assessments of climate sensitivity are derived from paleo data and from assessments of recent observations. These observations put a low end sensitivity of 1C. I don't really want to have an argument about whether these processes are good enough, but they cannot be dismissed in the way that you have done. I can see that you, Spencer and Christy are similar in that you all like your simple home-spun tales for how the climate should work and how things should be done, but it doesn't really reflect either the reality of the climate or the reality of how climate research is done. That's an easy one. Just because A can happen without B does not mean that A cannot also be caused by B. I think it's time to start a new climate model thread, don't you steve?
|
|
|
Post by steve on Aug 2, 2011 16:36:46 GMT
magellan,
I don't know. Checking back I've never started a climate model thread (apart from one I started in jest 2 years ago).
|
|
|
Post by magellan on Aug 2, 2011 16:38:21 GMT
magellan, I don't know. Checking back I've never started a climate model thread (apart from one I started in jest 2 years ago). I will then tonight if I have time. This will be fun.
|
|
|
Post by icefisher on Aug 2, 2011 17:56:25 GMT
2) His method of estimating climate sensitivity is so obviously faulty it doesn't really address other people's methods.
Where is your support for such a claim Steve? I can't respond to a claim you don't even specify much less provide any support for. But maybe its necessary for you to be inspecific lacking any specific complaint?
With my fading short term memory I am not sure where your interest in this even lies. Its my impression Spencer uses IPCC primary forcing assumptions for estimating sensitivity. So while there are issues there they are not issues one can criticize from the perspective of supporting IPCC estimated range of sensitivity.
After that there are only two routes for the heat generated by forcing. One is to radiate it back to space and the second is to send it into the ocean at an accelerated pace. Since all available evidence suggests the latter is not happening, radiating to space is the best science available.
So where is the beef?
|
|
|
Post by steve on Aug 3, 2011 8:57:24 GMT
icefisher,
His method for estimating sensitivity is based on an extremely simple model whereby heat is *diffused* into the ocean. The rate of diffusion, though, is a variable because we don't know accurately enough the rate of ocean heat uptake over a short period. Therefore the rate of diffusion can be modified to give you different values of climate sensitivity.
As far as I can see, and I've not looked in great detail, the Dessler paper is based on the assumption that clouds would be modified by atmosphere and sea surface temperatures etc. which are variables which we know reasonably well. Therefore any relationship between them could potentially be identified.
I believe that one of Spencer's criticism is that this temperature-causes-clouds is faulty, which is why he looks for faults in models. But as we've seen, the faults in the models have a better explanation (eg. that some have no ENSO). So even if he is right, this particular paper doesn't get us anywhere.
Just trying to understand things really. That's always been my interest.
|
|
|
Post by cybertiger on Aug 3, 2011 17:27:01 GMT
I believe that one of Spencer's criticism is that this temperature-causes-clouds is faulty, which is why he looks for faults in models. But as we've seen, the faults in the models have a better explanation (eg. that some have no ENSO). So even if he is right, this particular paper doesn't get us anywhere. I'm not very happy with the assumption that ENSO or other oceanic weather pattern oscillations have no long term climatic effect. They are very poorly understood, for example I know of no models predicting ENSO for more than 6 months and the existing models vary wildly in their predictions. I believe that was part of Spencer's criticism, the other part being that the existing models radiate heat out into space much slower after warming than satellite observations have shown over the last 10 years, which may or may not have been a consequence of ENSO, however it's still a valid criticism. For example two oscillations with a 10 and 11 year mean period might cause a 100 year global oscillation, and there's about 5 currently known and widely accepted oscillations which are very poorly understood. One of the biggest problems with models is that predict 50 years into the future based on the premise that it's correct for the last 10 years. Another problem is that the accuracy of global climate figures historically and that proxy records are only for a single location and/or are open to interpretation, we've only had satellite observations of temperature for the last 30 years or so, so it's a very hard to discount other causes. Even if AGW is a correct theory there is a very good chance I'll be dead before it's validated to my satisfaction, however I wouldn't want to say it's an incorrect theory at this point either.
|
|
|
Post by steve on Aug 3, 2011 17:58:20 GMT
It's all very well to say that oscillations *could* have an effect on the climate. The problem is that noone has a) explained why they should and b) demonstrated in any reasonable sort of way what the effect has been or indeed any good long term correlation. I would not be happy at assuming what the long term effects of ENSO were without evidence. If, for example, you look at the response of MPIs ECHAM5 model you find that the observations are comfortably within the window of the model's response (middle plot of the following) which somewhat undermine's Roy's criticism: In contrast with hypotheses about potential impacts of climate cycles, there is an easy to measure impact of "greenhouse gas" warming.
|
|
|
Post by sigurdur on Aug 3, 2011 18:16:18 GMT
Steve: The error bar is so large on each picture that you have posted that whatever the red and black lines are showing is really a flat line, or could be a flat line.
What am I missing here?
|
|
|
Post by cybertiger on Aug 3, 2011 18:25:57 GMT
It's all very well to say that oscillations *could* have an effect on the climate. The problem is that noone has a) explained why they should and b) demonstrated in any reasonable sort of way what the effect has been or indeed any good long term correlation. I would not be happy at assuming what the long term effects of ENSO were without evidence. I'm not happy assuming that they have no effect without evidence, which is what GCMs do, which was one of the issues Dr. Spencer highlighted. I'm not stating they have any effect, merely that you have to prove that they do not have any effect on long term climate in order to validate AGW, or alternatively accurately model them.
|
|
|
Post by steve on Aug 3, 2011 19:05:02 GMT
sigurdur,
It's not an error bar as such. It's a range of responses. In the real climate not all temperature peaks are the same. Ditto in the model climate. What the graph is saying is that the last 10 years of observations are within the bounds of the range of responses within 10 year periods of the model. It's trying to compare like with like.
I don't think it is quite fair to say that GCMs *assume* that El Ninos have no effect. El Nino's and their effects are just outputs of the model. Whatever Spencer highlighted, the ECHAM5 response suggests that he may be wrong.
|
|
|
Post by sigurdur on Aug 3, 2011 19:19:58 GMT
Steve: In effect, error bars. The range most of the time is over 4w....which is a huge amount of energy. I understand what the models are trying to express, but the average expression of R1 for all models really states nothing of certainty.
I think this is what Judith was expressing as an area that so far there are no credible results one way or the other.......
Correct?
|
|