ecs 2018 angech | February 6, 2018 at 1:37 am | Reply “There is also a puzzling peak below 1°C. These low values come from the GISS models (Fig. 7a) and if they are removed from the ensemble, the bump below 1K disappears . We find that 15 of the 25 CMIP5 models produce estimates in agreement with the CERES observations. If we limit the distributions to just those models , we obtain the ECS distribution in Fig. 6c (hereafter referred to as the “good” distribution). We consider the “good ” ECS distributions to be the best estimates of ECS from this analysis. Those ECS distributions have 17-83% confidence intervals (corresponding to the IPCC’s likely range) of 2.4-4.4 K ” ATTP has been running a series of articles on ECS, including a discussion of Marvel on 30/1/2018 and one discussing the “one-box energy balance model” by Clive Best by Mark Richardson 1712018 and a new one by Andrew Dessler technically Dessler and Forster) 4/2/2018. The gist of all the articles is that ECS is most likely 3.0 or higher with an inability or unwillingness to rule out much higher figures. – ATTP says “consider climate change specifically, then the no-feedback response is about 1.2K (i.e., the no-feedback response to doubling atmospheric CO2). This is largely because the Planck response is 3.2W/m^2/K and the change in forcing due to a doubling of atmospheric CO2 is 3.7W/m^2.” – This issue is very important as shown by the time and effort put into denigrating lower estimates like Nic and Judith’s. The current trend is to blame the observations for showing lower climate sensitivity than the models and then using the models to prove it should be higher. – Basically a lot of the AGW concern falls over if ECS is 2.0 or less hence the concerted effort to deny this.. – Andrew Dessler has an interesting take on using short term observations 2000 to 2017 to achieve an estimate that fits the models. The only problem is a Gerghis like selectivity of the models he wishes to use for his Monte Carlo runs. 2, based on GISS, suggested ECS in the 1.0 or less range. Fortunately these were not needed for the 15 out of 25 model ensemble used showing an ECS of average 3.3. – Nonetheless for ECS fans, good reading, and an excellent counterpoise to ideas here. niclewis | February 6, 2018 at 3:50 am | Reply Yes. The GISS models are certainly outliers is various ways. But that doesn’t necessarily mean that what they imply about ECS is wrong. It has, incidentally, always struck me how different Gavin Schmidt’s views on ECS are with the behaviour of the GCMs developed at the GISS institute that he heads. The biggest problem with Andrew Dessler’s approach is that, as he states in the paper, “the transfer function ?_IV/ ?_4xCO2 seems the most probable place for a significant error to occur” and they “have no way to observationally validate it, nor any theory to guide us”. That transfer function is the scaling factor they use to converting observations of short-term, mainly unforced interannual climate variability to an estimate of the response of the climate system to long-term forced warming. It is the biggest contributor to uncertainty in their ECS estimate. They estimate the transfer function using GCMs, but if GCM’s high ECS values are mainly the result of their wrongly simulating long term cloud feedbacks then is is highly probable that their values for the transfer function will also be systematically wrong.Interesting post on many levels hence the large number of comments. Several hit nerves. Susan mentioned the unbelievable support for Trump. One of the mysteries that might deserve it’s own post one day in terms of whether this effort by John is helpful or not as Joshua mentioned. The similarities in presenting 100 fallacies in Climate change and presenting 100 reasons why Trump should not have been elected share an end result. The unbelievable happens. Perhaps giving 100 reasons is trying a bit too hard? The end result of trying to hard is that people suspect one of having ulterior motives? Or as might be surmised the BS meters tend to go off. They look at the arguments and say if they are trying so hard to convince otherwise perhaps there is something there. It is just a thought. – I take this as a good summary of the the differences in our points of view. By putting 100 reasons out there, in their 5 groups, all supporting each other there is a unified purpose and consensus for all to agree on. No I will have a little bit of this and all of that. It gives skeptics a good starting point as well to review their positions but most importantly it nails the colours to the mast. If any one point turns out to be right for the skeptics then all points will be adjudged right for them. Best put the other way round. If just one of John’s premises were to fall over, the rest, even if right, become shaky and may fall. I wonder which will be first?