|
Post by steve on Aug 4, 2013 13:37:32 GMT
You also need to understand that the ONLY reason the atmosphere can cool is due to the presence of CO2 and H2O radiative gases. CO2 molecules scatter incident infrared in three very narrow bands; if they collide before re-emitting incident infrared they can pass on that energy to the colliding N2 or O2 molecules. If a CO2 molecule is given energy by collision it can radiate that energy as infrared. H2O absorbs heat from infrared and from collisions and may release it on state change. H2O molecules can also change state on absorbing heat without changing temperature and release heat on state change without changing temperature. Infrared released by H2O on state change (condensation and freezing) is not subject to Stefan Boltzmann radiation equations. Liquid water can exist up to 30,000 ft or more in convective updrafts. The simplistic 'greenhouse' effect takes no account of the above facts. Further simplifying the description of the 'greenhouse' effect adds even more simplification errors and discrepancies. It depends on your reading of the "simplicity". For example: If you take a cold atmosphere and irradiate it with IR, then the CO2 and H2O molecules will absorb some of the radiation then collide with O2 and N2 molecules. This will cause the atmosphere to warm up. You could say that this is due to the heat being "trapped" by the CO2 and H20 molecules because of the *fact* that adding more CO2 and H2O will tend to warm the atmosphere more. The fact that they act as a trap is dependent on the fact that the profile of the atmosphere is warm and dense at the bottom and cold and thin at the top is for the purposes of the explanation neither here nor there. Regardless of the semantics, though, estimating the effect of adding CO2 to the atmosphere is done by sophisticated models that do not have the arguably erroneous simplifications.
|
|
|
Post by steve on Aug 4, 2013 13:26:46 GMT
Couple of comments: 1. While the insulation is perfect, no internal source is required to keep the inner shell to the temperature required to maintain 400W/m^2. It emits 400W and receives 400W.
2. Once you start to strip away the insulation, the outer layer will start to lose heat to outer space, so will cool. As it cools, it will radiate steadily less than 400W inwards. Therefore the inner layer will *also* start to cool *unless* you turn on a heat source to maintain the inner layer at 17C. You will have to gradually increase the heat source as the insulation is removed if you want to maintain 400W.
If the outer layer is reduced to a thin layer of glass then: - either it is completely transparent in which case it neither absorbs any of the 400W, and nor does it emit any (Kirchoff's thermal radiation law - so essentially this is an impossible material). - or it absorbs a small amount of the energy which will warm it such that it will be able to emit the energy too - some outwards and some inwards.
|
|
|
Post by steve on Jun 19, 2013 6:09:26 GMT
Icefisher,
I'm not saying funding hasn't been sufficient in the past but the problem is growing because the computers are getting more expensive to buy and run and because the problem is getting more complex. Part of the increasing complexity is down to the better observation systems that are in place which put the model developers under more pressure a) to build in better simulations and b) to match more detailed observations. Which leads to your other comment.
There are three reasons why a bigger computer will help: 1. Higher resolution of the same model, 2. ensembles - running the same model lots of times but each version with a tweak helps give an idea of its variability and range of response - so you are less dependent on one outcome which may have been the extreme response. 3. including more science (chemistry, carbon cycle etc.)
There is a Tower of Babel issue with explaining the results to folk like us - but it is a complex issue anyway so you shouldn't shy away from it. The individual people working on the individual aspects will get more opportunity to understand more about their particular issue. In the fable, the individual groups of people with their individual languages have ended up building multiple competing "Towers" to look into the mind of "God" (LHC, GCM, HST, JET, Curiosity, Voyager etc.)
|
|
|
Post by steve on Jun 18, 2013 5:46:33 GMT
No far reaching political decisions have been made on the strength of the seasonal forecast. I think unwary politicians should be chucked out of office. They aren't all stupid, and they obviously make decisions based on many other models, theories and gut instincts which have less basis than a seasonal model. There are plenty of politicians who will make much of the colder wetter summers when the forecasts are predicting drier ones (in 20-30 years mind you). That said, yes they should be wary. I understand the later models are showing some parts of Europe getting wetter summers, not drier ones. That's why we need better and higher resolution models which means much bigger computers. Only minor things like the amount of grit and salt or the number of snow clearing machines. But what's a few deaths in the snow and closed airports? And as the Met Office has been proudly telling us the same model is used for both short range and climatic forecasts so the models have been used to good effect look at 4200 windmills - only there because of the Met Office model convincing parliament in 2008. I can't remember anyone talking about probabilities and error bars. And of course the Met Office was telling politicians with inside trading interests precisely what they wanted to know. So no - nobody is taking the Met Office seriously. Well not now. Nautonnier, Heathrow was closed in 2009(?) because they didn't invest in equipment. At the same time Heathrow was closed, Gatwick which is just down the road was open and the M25 between M3 and M4 less than a mile from Heathrow was clear and driveable at full speed. I know this because I flew into Gatwick while colleagues flying to Heathrow were stuck. Blame the Spanish owners, not the weather forecast. But this conversation is getting confusing. The original issue was whether the Met Office would be thrown under a bus because a high-profile but relatively minor output - the seasonal forecast - was embarrassingly useless. It won't be. I'm pretty sure that the ministers with the purse strings are well aware of the uncertainty, and previous and current government chief scientists as well as other high profile scientists have banged on pretty loudly about the need for proper funding of climate and seasonal research.
|
|
|
Post by steve on Jun 18, 2013 5:33:11 GMT
Objectively that story sounds rubbish. I haven't been following this but it sounds like a London-based journalist has got a bit hot under the collar because a bit of warm air has pushed across the Channel. The forecast has been and continues to be for changeable weather with average to cooler temperatures more likely. I've been keeping an eye on it because I'm looking forward to my second wet and windy Welsh holiday in two consecutive Junes. One or two days of slightly warmer temperatures in London doesn't change or undermine this forecast especially when it is qualified with: "it it will not be wall-to-wall sunshine; it will be mainly cloudy with a chance of thunderstorms".
|
|
|
Post by steve on Jun 16, 2013 17:34:48 GMT
The seasonal forecast is not a bread-winner for many agencies since few models have skill in general and none have much skill over NW Europe due to the fact that the weather is balanced between very different continental and ocean types. I understand that the recent improvements in resolution are hinting at better winter skill, but summer skill is not there (for anyone). The government will throw money at seasonal forecasting until it can be shown either that the problem is unsolvable or that the Met Office are incompetent compared with anyone else, and I've not seen evidence that commercial forecasters do it better. I've seen plenty of BS from certain such forecasters who depend on predicting dramatic weather non-stop and then making a big thing about their "success" when occasionally dramatic weather occurs. Most proper companies are dependent on model outputs and there isn't a huge differential between the major models, though I believe ECMWF has the edge for seasonal forecasts. With its vastly larger piles of tax payer cash, its huge EU-gravy train salaries and its more limited remit it ought to. It is unfortunate then that so many far reaching political decisions are made on the strength of the 'seasonal forecast'. Perhaps the Met Office should make it plain that these are really experimental forecasts with doubtful skill and indeed have only been correct less than 1 in 10 times (with all the errors in the warm direction). Therefore, long term decisions on water policy, winter snow clearance and energy requirements should not be based on the Met Office forecasts. I cannot remember seeing this type of caveat on any of their forecasts although the use of probabilities does allow them a mathematical escape (we said it was only a 5% chance not impossible) that the unwary politicians will not understand. No far reaching political decisions have been made on the strength of the seasonal forecast. I think unwary politicians should be chucked out of office. They aren't all stupid, and they obviously make decisions based on many other models, theories and gut instincts which have less basis than a seasonal model. There are plenty of politicians who will make much of the colder wetter summers when the forecasts are predicting drier ones (in 20-30 years mind you). That said, yes they should be wary. I understand the later models are showing some parts of Europe getting wetter summers, not drier ones. That's why we need better and higher resolution models which means much bigger computers.
|
|
|
Post by steve on Jun 16, 2013 15:06:06 GMT
Indeed you don't throw the weather agency under the bus. But a major reorganization would not be included in that prohibition. When it comes to climate prediction they are at least strike two and not close to getting a walk as thus far the agency has provided no tangible value and has caused harm both from above and from the botched seasonal forecasts a few years ago before they went underground with that trash. A slightly delayed tanker from the middle east and there would have been power cuts in March. Since then they have closed a couple more power stations with no replacement. The Met Office does not publicly call an urgent meeting about its systemic lack of skill over 12 years without there being panic in the system (steve is probably closer to that then I am). They make another goof like 2012 and there will be cuts. Especially as their long range forecasting showed less skill than commercial counterparts that cost the taxpayer nothing. All the Exeter empire building seems to have resulted in a less skilled organization. I am not suggesting that they move back to Bracknell but they could easily be privatized as DERA became Qinetiq, cease to be supported by the exchequer and have to bid against other weather companies for government contracts. It is extremely unlikely that they could continue their work on the 'world stage' in climate science as they can only do that with government funding and they would need to bid for that as well. For a long term Met Office employee that is going under the bus. Put the cost and lack of skill demonstrated by the Met Office against the self funded skill of commercial weather companies and the Met Office would be unlikely to be contracted by any sane contracting officer. The UK finances are not in altruism mode at the moment. The seasonal forecast is not a bread-winner for many agencies since few models have skill in general and none have much skill over NW Europe due to the fact that the weather is balanced between very different continental and ocean types. I understand that the recent improvements in resolution are hinting at better winter skill, but summer skill is not there (for anyone). The government will throw money at seasonal forecasting until it can be shown either that the problem is unsolvable or that the Met Office are incompetent compared with anyone else, and I've not seen evidence that commercial forecasters do it better. I've seen plenty of BS from certain such forecasters who depend on predicting dramatic weather non-stop and then making a big thing about their "success" when occasionally dramatic weather occurs. Most proper companies are dependent on model outputs and there isn't a huge differential between the major models, though I believe ECMWF has the edge for seasonal forecasts. With its vastly larger piles of tax payer cash, its huge EU-gravy train salaries and its more limited remit it ought to.
|
|
|
Post by steve on Jun 15, 2013 14:52:22 GMT
The Met Office are about to be thrown under the bus. They have done their part in providing the ammunition for setting up the tax regime and crippling the energy supply. UK will run out of electricity generation capacity inside 18 months thanks to the 'green' advice that the Met Office is now admitting was "a systemic error", but which led to the Climate Change Act and the associated stupidities and troughing greed of the MPs. Imagine the effect of a severe winter with rolling power outages. Plan for how you will deal with extended power cuts because with the continuing closure of baseload power generation and no replacement that is what is going to happen. I suspect that there will be no such power cuts in the UK and that the Met Office will be thrown bags of money rather than be thrown under a bus.
|
|
|
Post by steve on Jun 13, 2013 7:00:15 GMT
Icefisher, Santer has written a lot of stuff, so I'm not sure what you are referring to. I am talking about, and told you a while back, the work of Santer in AR3 that he now continues to defend with many writings against challenges against it as if it were the null hypothesis. I remember you mentioning AR3. AR3 is about a thousand pages long. Can you be a bit more specific because there seems to be a narrative going on here that I'm probably unfamiliar with. Here's the link to AR3: www.grida.no/publications/other/ipcc_tar/
|
|
|
Post by steve on Jun 12, 2013 12:10:47 GMT
Icefisher,
Santer has written a lot of stuff, so I'm not sure what you are referring to.
|
|
|
Post by steve on Jun 11, 2013 20:20:43 GMT
Icefisher, A periodogram of 170 years of data *cannot* show a *significant* cycle that is 60 years in length! It is patently obvious that such a series can show at most 3 such "cycles" which is not significant. It's also obvious that many random or natural series with random influences will show cycles of varying lengths. You only have to watch the tide come in for half an hour to see an example of such cycles. I'm not hung up on the exact length of the multidecadal global temperature cycles although I think for purposes of predicting future cycles, 60 years has a reasonable historical basis. It does,however, lack an explanation for its existence and as Icefisher notes Santer does provide an explanation for 70 years although it is very controversial. If you want to use 70 years, then my prediction for flat temperatures would extend to 2047 rather than 2037 as I now predict. It's much more important that one understands that there have been multidecadal trend fluctuations in global temperatures and that this is related at least in part to Ocean Cycles and then incorporate this knowledge into future predictions. Many, if not most, of the warmists concede that Ocean Cycles impact global temperatures significantly, but they just wring their hands because they "don't know how to model them". The result of this omission is totally useless models which give highly inaccurate and therefore misleading predictions. This is something that can be easily improved and needs to be improved but the incentives may not be right. I think climate scientists accept that the ocean can significantly impact the trend. Whether it is thought to be a result of "cycles" or whether it is more random than that, though, is perhaps another matter. I don't think it is fair to talk about them "wringing their hands" when the alternative solution is simply to look at the past and assume that the future will be the same. It is a hard problem and even a perfect model would fail due to lack of good observations with which to initialise the model (as well as the fact that even the most powerful computers cannot effectively simulate an ocean to more than about a tenth of a degree resolution). If you turn out to be right, it simply means that rapid warming will start again in 2037...
|
|
|
Post by steve on Jun 11, 2013 12:17:49 GMT
Icefisher, A periodogram of 170 years of data *cannot* show a *significant* cycle that is 60 years in length! It is patently obvious that such a series can show at most 3 such "cycles" which is not significant. It's also obvious that many random or natural series with random influences will show cycles of varying lengths. You only have to watch the tide come in for half an hour to see an example of such cycles. What you can show Steve is the potential for such a cycle. . . .meaning you can not rule it out by putting a bag over your head and pretending such natural variation does not exist when clearly it does. It is not a significant feature, and there is no good physical reason for expecting such a feature. Therefore focussing on a 60 year cycle seems misguided to me. That's not ruling it out. It's setting priorities and likelihoods.
|
|
|
Post by steve on Jun 11, 2013 6:29:59 GMT
Prof. Murry Salby’s Presentation In Hamburg: Climate “Model World” Diverges Starkly From “Real World”. "Prof. Murry Salby, climate scientist at Macquarie University of Sydney, made a presentation in Hamburg on April 18th as part of a European tour. Prof. Salby is author of the textbook Physics of the Atmosphere and Climate (Cambridge University Press) and Fundamentals of Atmospheric Physics (Academic Press) and is renowned worldwide as an astrophysicist." Model world (left) vs real world (right). The points of Salby’s presentation lead to the following implications: - In the Real World global temperature is not controlled exclusively by CO2, as it is in the Model World. - In significant part, however, CO2 is controlled by Global Temperature, as it is in the Proxy Record. Report from notrickszone : notrickszone.com/2013/06/10/murry-salbys-presentation-in-hamburg/ cuttydyer, You have to remember that the model vs reality plot is skewed because so far we only have data for a tiny section of the plot on the left. With regard to temperature controlling CO2, the amount of CO2 emitted by man can be reasonably estimated, and amounts to 2-3 times more than is required to cause the observed rise in CO2. CO2 rise is moderated - probably it is mostly by absorption of CO2 by the oceans. As well as measuring man's CO2 by looking at coal, oil and gas burning statistics, you can also observe the reduction in levels of oxygen that is due to the combustion of such fossil fuels. The amount of oxygen in the atmosphere is reducing (very slightly) and the reduction is in line with the increase in CO2. www.ipcc.ch/publications_and_data/ar4/wg1/en/ch2s2-3.html#2-3-1Given that the ocean is an important sink for CO2 and given that warmer waters absorb CO2 less effectively, the sea surface temperature can moderate the amount of absorption - and in this way can effect levels of CO2 in the atmosphere.
|
|
|
Post by steve on Jun 11, 2013 6:09:41 GMT
Icefisher,
A periodogram of 170 years of data *cannot* show a *significant* cycle that is 60 years in length!
It is patently obvious that such a series can show at most 3 such "cycles" which is not significant.
It's also obvious that many random or natural series with random influences will show cycles of varying lengths.
You only have to watch the tide come in for half an hour to see an example of such cycles.
|
|
|
Post by steve on Jun 11, 2013 6:02:26 GMT
Duwayne, The reason was to give both trend lines more than 30 years of data. But you get the same result with 1962, 1972, 1977 and pretty much anything else I've tried before 1980. At 1982 the trends are pretty similar as the relative importance of the 1998 end-point increases. Steve, if any starting point before 1980 is equally good for the case you are making, then I'd suggest you use 1977 as the beginning point because it's the year when the Ocean Cycle reversed to its warm phase. And I'd suggest an end year of 2007 since that's when the Ocean Cycle reversed to its cool phase. Others may give slightly different years, but these seem to be the best to me. Now, I realize this doesn't overcome the problem with a purely statistical analysis of recent trends to determine if temperatures have flattened. But scientists should try to wring as much information out of the data as they can realizing there is no more data available. So, I've proposed that one first determine the trend value for the year 2007 based on the 1977 to 2007 data. Using Hadcrut3 this is 0.47C. I don't know of any better way for establishing a benchmark against which to compare future temperatures to determine whether they are rising, falling or flat. As you said above there are 30 years or 360 months of data which went into this calculation. Selecting a year later or a year earlier changes the calculation by only 0.01C. If temperatures remain flat for the 30 years from the 2007 starting point, then the average temperature should be about 0.47C. So far we are 20% of the way there and the average for 2007 through 2013 is 0.40C. That's unfortunately all the data that is available. So the question then becomes - Is there any reason why these data might be providing misleading results? Some might point to the fact that ENSO and the PDO have been negative during this period and are causing the temperatures to be lower than normal. That's true. But The ENSO and PDO values are almost exactly what they were during the last cool Ocean Cycle period and are exactly what one would expect. That in fact supports the argument for why temperatures are flat and the Ocean Cycle theory in general. Meanwhile, the 11-year Solar Cycle was on the warm side during much of this period and is currently near its maximum. It is clearly not causing temperatures to be lower than normal. And there are no large volcanic eruptions pushing temperatures down. It's clear to me that the data show that global temperatures from 2007 until now are not going up and may actually be cooling and this is in line with expectations from the Ocean Cycle. I don't think you can pick any one, two or three years. If temperatures remain the same as 2007 for 30 years, then that will be telling - it will also reduce the trend from 1970-2043 by a half. But I don't think you can say anything meaningful about the trend from 2007 for 10 or 20 years (by looking at the trend). The way to get more meaning out of the data is to examine the system in more details - we need to continue and improve the observations we are taking so we can see where the energy is coming from and going to. And we need better models to help us understand the observations.
|
|