Reto and Peter, ECMWF is the best forecast model in the world. If you had attended the change of directorship last Spring, you would have heard them say that the ice model predictions ( and they have the HIGHEST resolution forecasting model) are terrible, do not predict changes and have little connection with reality. My climate model has taken 250 days on 12,000 cray cores and I only achieve 20km resolution. ( I estimate I used about 2 million dollars of electricity in the end, as some of the model is at 2-3km resolution. It has possibly been one of the biggest projects on ARCHER.
ECMWF can run globally at 2-4km resolution. How is it also that Martin Miller , who masterminded much of the ECMWF model said of AR5 , that the clouds were so badly predicted that the radiative balance is not correct, and the climate models could not be trusted for clouds and radiation? He was one of the worlds best experts,but ignored. Also then, How is it that then the climate models can do so much better than the best weather prediction models in the world? When the data evidence of the HUGE changes in ice volume and extent, are so apparent, so much so that by the time AR5 went to press , six months after writing, the ar5 ice extent predictions were outside the predicted range for the worst scenario, how can the ar5 community still try to sell the story? Maybe Trump took his lead from the climate modellors? I have had endless arguments with Piers Forster , another lead author on ar5 about this. scientists who know about modelling the weather have been ignored. You do not get papers published which day that models are wrong, only when you say how great they are. In essence ,after AR3, it has become a gravy train for academic careers based largely on CXXP models. In a short time these models will be shown to be catestrophically wrong, and all that is done is to pillory those who try to point out the true science. Regards Alan. T --- Alan Gadian, NCAS, UK. Email: [email protected] & [email protected] Tel: +44 / 0 775 451 9009 T --- > On 23 Nov 2016, at 15:32, P. Wadhams <[email protected]> wrote: > > Dear Dr Knutti, I would like to intervene in this seemingly interminable > thread to draw attention to errors which you have made when you mention my > work. > > 1. " But the natural variability is very large (e.g. Kay 2011, >> Swart 2015, Screen 2013, 2016). There is no reason why short term trends >> could simply be extrapolated, and the predictions by Peter Wadhams of sea >> ice disappearing by today have not happened so far" > a) I completely agree that natural variability is very large and stress this > in my book "A Farewell to Ice" which I suspect you have not read. In the book > I stress that it is the TREND which is rapidly downwards towards zero, not > that every successive year features a monotonic decrease in summer ice > volume. And, despite the newspaper reports which seems to constitute your > reading matter on my work, I do NOT forecast that the ice volume will go to > zero in a specific year (like this year), only that it will go to zero in a > very small window of years (of which this year forms one) as opposed to the > decades which are still forecast by the worst of the current models. You > might also care to glance at the current NSIDC charts for ice volume (Arctic > and Antarctic) which show an extraordinary turndown since September > suggesting that the co-operative collapse which I have long projected through > my thickness measurements is now happening. b) Why can't short term trends be > extrapolated? Firstly, they are not short term - they date back to the 1980s > when observations began. And secondly, please name the magical effect which > will cause the trend to become invalid and the ice volume to pick up again. > > 2. > Maybe the models are missing something, but it's just as plausible (and >> in most scientist's view more likely) that the models are largely consistent >> with observations within natural variability. > Your idea of "most scientists" is an odd one. When the observations show a > rapid declining trend leading to zero very soon, how can models (which > predict decades of continued summer sea ice) be consistent with these > observations? They are simply not consistent, and it is clear that by "most > scientists" you mean "many modellers", some of whom have never seen sea ice. > > 3. I'm not >> downplaying the strong changes in the Arctic, but the science suggests a >> fairly linear (and reversible) relationship between Arctic sea ice and >> temperature with large variability superimposed. In my view they do not >> support a "death spiral". > Well, it's obvious that there is a linear relationship between air > temperature history and sea ice thickness. The "degree days of cold" analysis > dates back to the 1950s and was amplified by Maykut and Untersteiner's > analysis in 1971. But if air temperature is increasing rapidly, as it is, > then thickness will decline just as rapidly, leading to a period in summer > with zero thickness, in other words, a "death spiral" QED. > > To paraphrase Captain Scott, I don't think I need say more, > Best wishes > Peter Wadhams > > > -- You received this message because you are subscribed to the Google Groups "geoengineering" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/geoengineering. For more options, visit https://groups.google.com/d/optout.
