Discussion > Do computer models provide 'evidence'?
There was an interesting program on the TV channel Quest, 19:00 on Sunday night by João Magueijo a cosmologist who explained the problems with the existing theory about the big bang.
He has proposed a new theory which requires the speed of light to have been much faster just after the big bang, this goes against accepted wisdom, that it is a constant.
In conclusion he said that it was up to him to find the observational evidence to show his theory is correct, and that this is the basis of "all science". Not computer simulations of what people think might be happening but real world evidence.
It's a pity that climate 'scientists' don't adopt the same approach.
Reading this article it seems that cosmology and climatology share similar problems with a lot of vested interests trying to protect their position and 'life's work'.....
http://www.csmonitor.com/2003/0130/p14s03-bogn.html
Another attempt to explain to Raff why models aren't evidence.
Say a murder takes place. There are all sorts of evidence from the scene and the suspect. That is stuff that can be observed. After that the prosecution creates a scenario to draw the evidence together. Climate models are not the evidence, they’re the case for the prosecution, sometimes called supposition. Real evidence forms part of them, but they are not evidence themselves. To counter the prosecution there is the defence. It too has physical evidence and it casts doubt on the accuracy of the climate model picture of events.
At this point the climate debate and criminal justice parts company. To counter the holes in the climate prosecution case, warmists demand that the defendant find the real culprit or failing that, ask the jury to find CO2 guilty just in case.
TinyCO2
Unless you've persuaded raff to modify his view since the start of this thread he is (was) arguing that computer models could be used as evidence that the murder was going to happen provided they had been more or less correct in predicting such events in the past.
Twice I've made the point that you cannot have "evidence" for something that has not yet happened. You can have predictions on which you may be able to rely up to a likelihood of about 95% or where all parameters are known even close to 100% provided the models are good enough but even there what you have is not evidence and cannot be in something which is as unpredictable as weather and about which we know as little of the working as climate,.
Which is why I challenged him the othe day (a bit tongue in cheek) to model next year's Epsom Derby. All the parameters (probably more than modellers use for their climate forecasts) are available though a little tweaking would be allowed up to a week before the race. But there is no way that the output could be described as evidence.
Mike Jackson "Twice I've made the point that you cannot have "evidence" for something that has not yet happened. "
True but models are also used to explain the past and even that isn't proved.
"Twice I've made the point that you cannot have "evidence" for something that has not yet happened. "
What more evidence do you need that you are wasting your time?
You know me, I always think every believer troll is a potential save. I was partially correct with BitBucket, and so it may be so with Raff. He's doesn't seem stupid (just deliberately obtuse) so I think he's worth working on.
How about if raff, are you open to persuasion?
Raff: you seem to insist that models can provide evidence. I, and others on here, argue that the only evidence that can be gained is whether or not the assumptions and inputs into the model are correct. What other evidence do you think is obtainable from models?
Am I open to persuasion? I told you before and only TBYJ has had the honesty to accept my invitation (for which I salute him). To repeat again, I will accept that models produce no evidence if each of you admits without artifice that there is no evidence that any degree of warming will be beneficial and that people like Ridley and Montford are completely wrong to claim that it will be or to draw attention to model-based research that suggests it will; and that there is no evidence that taxing carbon emissions will cause great economic harm...
And for good measure, I want an explanation, without support from any sort of model, for these 'facts' (they were stated as facts):
-- when it gets hot the effect of H2O is to redistribute the energy faster (presumably that means faster than if there isn't any water).
-- cumulus clouds are negative feedback (presumably the suggestion is that warming produces more cumulus clouds and that these reflect sunlight etc).
Raff it doesn't seem we even agree with you as to what constitutes evidence, so there's no point going into the complicated stuff.
Why should we say that there is no evidence that any degree of global warming will be beneficial, whether we use artifice or not, when there is ample evidence that it will? Fewer winter deaths from hypothermia; increase in growing season; increase in crops per land area; greater area of arable land – indeed, greater greening of the continents in general. Do you really expect us to lie to placate you?
For the final points of your missive, you are now showing the use of models – they can demonstrate that the claims that you make are either right or wrong, depending upon the suppositions (the information) you enter. That is the ONLY evidence that they can provide! (A point, I note, that you studiously avoid addressing).
If you think that models can produce other evidence, please tell us what this evidence would be.
Raff
The evidence that we have suggests that the human race prospers more in warm climates. However I accept that
1. There is no computer model used by the IPCC to confirm this
2. We have to rely on proxy data to confirm that there were Medieval Warm Period, Roman* and Minoan* Optimums.
3. We have to rely on proxy data to confirm that these were global
4. We have to rely on proxy data that the global temperature was greater than today
5. We have to rely on proxy data to confirm that civilisations and humankind flourished at these times.
6. We have to rely on proxy data that the Little Ice Age was colder than today
7. We have to rely on proxy data to identify other cold periods during human history
8. We have to rely on proxy data to confirm that civilisations and humankind don't thrive and prosper during colder time
9. We have to rely on proxy data to confirm that civilisations decline and disappear in longer colder times.+
*So called because the Roman and Minoan civilisations thrived at these times.
+ The Sea Peoples invasions of Egypt and the eastern Mediterranean coincides with a climatic cooling about 3500 years ago
+ At the end of the Roman Optimum ~400 CE the Vandals moved from northern Europe through France and Spain to North Africa eventually sacking Rome.
+ Beginning around 440 C.E., the Huns initiated a series of invasions and gained control over north western India. Unable to restore control over even its tributary princes, the Gupta dynasty collapsed entirely around 550 C.E.
+ Sahara Desert, about 3,500 years ago the Saharan lakes finally dried up, the vegetation disappeared, and the people vanished.
Like the "Head on the railway track experiment" (the counter arguments provided you have ignored) we cannot be certain what the future holds but the evidence we have suggests that if the climate warms then overall the human race will be better off.
I would regard this as a non-model confirmation that using what historical evidence we have that if the climate continues to return to previous warmer levels then the most likely outcome is an improvement in the lot of the human race in general; although there is a possibility there may be localised areas which lose out.
I submit that you are completely wrong and unaware of the evidence which strongly suggests that a warmer climate will be beneficial.
Now I challenge you to produce non-model evidence that I am wrong on this and on your head on railway line experiment.
Raff is trying to get the CAGW hypothesis accepted as the null in his set of bogus questions. I am not playing. My own null is that nothing much is happening and we can adapt if it does. I am not inclined to accept the other side's assertions as true. They need to be proved. Proved in the old-fashioned sense, tested against real observations past and present. It is the tendency of the climare crowd to skip this bit and the tendency of some of them to cheat their way through it, that makes me doubt the whole shebang. The way their proposed solutions always seem to be counter-progress (as I understand it) also causes suspicion.
Rhoda,
If we take the two possible extremes of climate on Earth, Snowball and Venus, then neither will be good for most forms of life. Therefore between the two there will be one or more optimums for various carbon based lifeforms. From what we see around us over a range of sub-climates is that fewest people live in the extremes, polar, desert, mountain and remote islands. The population is most vulnerable in these areas. Temperate zones support life quite well most of the time but prolonged excursions to more polar or more dry (less wet?) can cause major issues. Wet tropical zones don't seem that ideal either as the population in the rainforests of the world may be diverse but they are not that dense when compared to the grasslands of Africa or pre-Colombian Northern America.
It seems to me that, for Europe and probably globally the LIA was a hard time in survival terms, MWP was a lot easier. The goal of "reversing climate change" which really means cooling the planet to 18th-19th century levels seems at the very l;east misguided.
Having said all that my position is: all the evidence is that the climate is within all previous extremes of all measurements/proxies in the current epoch of glacial/inter-glacial and therefore cannot be considered to be changing. Based on the previous two paragraph summary a slightly warmer climate would enhance the well being of humankind. An ice-age or even a major cooling, which is still the most likely long term future would not be good.
Basing planning on unproven climate science and models and totally ignoring the past makes little sense and is likely to lead to major problems. The great Xhosa Cattle-killing is probably an extreme example.
Not just prejudice of the programmer, the ignorance as well.
Radical, the world is a big, big place. Most items on your list of benefits apply only to the colder, richer parts of it. Most of humanity doesn't live in those parts. What about the rest?
Sandy, you may be a historian in real life, but I doubt you'd get far with a hypothesis that these civilizations rose and prospered *because* of the temperature. Is it not equally likely that things happened *despite* the temperature or that temperature was really not a major factor? Can you make a case that Chinese or South American civilizations were most successful at times of greatest temperature? I very much doubt it. You have just picked a set of historical data that suits your desired story. Modern industrial and post industrial societies developed in the cold north, not the warm Mediteranean; do you think the prospect of still warmer weather in Mediteranean countries would be welcomed.
To show that warming is beneficial you have to consider the world, not your back yard. I doubt you can do this with any rigour using historical anecdotes.
rhoda, what would you need to see to conclude that your "nothing much is happening" is untrue?
re TinyCo2 and "Raff it doesn't seem we even agree with you as to what constitutes evidence, so there's no point going into the complicated stuff."
If you go back to page 1 then you will see that Raff starts his/her contribution with the following definition taken from a dictionary.
"I have seen various definitions of 'evidence', including:
- A thing or things helpful in forming a conclusion or judgment
- Something indicative; an outward sign
- Ground for belief or disbelief
Model output seems to fit each of these. That is not to say it is 'proof'. Just like lines of evidence in a court case may be
false or may still point to the wrong verdict. Models are tools, they serve a purpose. They are not perfect and their outputs need to be interpreted and treated with caution. But that doesn't prevent them from being useful. If you reject models, what do you propose in their place, if (although this seems unlikely) you want to understand the likely course of the climate."
I think it is necessary to make the observation that there is a clear distinction between this and "scientific evidence". For example see Wiki.
http://en.wikipedia.org/wiki/Scientific_evidence
"Scientific evidence is evidence which serves to either support or counter a scientific theory or hypothesis. Such evidence is expected to be empirical evidence and in accordance with scientific method. Standards for scientific evidence vary according to the field of inquiry, but the strength of scientific evidence is generally based on the results of statistical analysis and the strength of scientific controls."
I am quite clear that the original quote "“Lord Lawson’s views are not supported by the evidence from computer modelling and scientific research.” whilst not explicitly stating that computer models are scientific evidence is putting them on a par with scientific research and implies they provide that level of evidence. On that basis I was quite happy with the original post by Martin A which is concerned with the use of climate models as supporting/representing a scientific hypothesis and the need for validation against emperical evidence.
On the other hand Raff uses the alternative defintion. Again I am quite clear that Computer Models can be evidence of things. For example climate models are evidence that people have written a lot of code or that they have been able to model a hypothesis.
What has bothered me about Raffs posts is that they started from the outset in using the alternative definition and then effectively failed to engage on the sustantive issue raised by Martin A of why models that fail to validate should be treated as scientific evidence.
Raff: you asked about evidence of benefits of global warming; I offered you a few, which, apparently, now have some codicil attached that invalidates them – and you have accused others of moving the goal-posts! (However, you have, perhaps inadvertently, given us an interesting insight into your mentality.)
The simple fact is that the warmer areas of the world have not registered any significant change in average temperatures – indeed, one of the planks of this entire AGW/ACC scare is that the warming is concentrated in the higher latitudes, so we should be scared as all that ice will melt! However, the whole of this big, big world is greening; probably less to do with warming and more to do with rising CO2. But – as the two are supposed to be somehow inextricably linked – then it is one of the benefits of warming.
Now, let me restate an observation that you do seem reluctant to address: the ONLY evidence a model can provide (apart from the peripheral evidence Clivere refers to) is whether or not the assumption made in its creation were correct. Should you know of any other evidence that can be derived from a model, please tell us about what this could be, and shed a light on our woeful ignorance.
It is odd that people who challenge models are also such keen users of model results. There was Rhoda with her certainty that cumulus clouds are negative feedback in action and that energy is distibuted faster (than something unspecified) by water in the atmosphere. Now Radical uses models against me. Warming is supposed by models to be concentrated towards the poles and therefore my argument that the effects of warming across the whole globe not just the colder areas need to be considered can be rejected. There is no warming in hotter areas he says (and by inference, there will be none) and so they can be ignored.
Yet if I look at spectral maps I don't see no warming in the hotter parts. And global warming is unlikely to leave regional weather systems of hotter parts of the planet unaffected. So you do need to consider what might happen in hotter, poorer places. They might get greener, but that doesn't in itself represent an economic gain - the devil is in the details.
Raff,
rather than come up with counter evidence you took the cowards way out, "but, but you haven't mentioned EVERYTHING".
But seeing as you ask, The Golden Horde, decline starts ~1350 LIA about 1300 to 1870
The Ming dynasty, also Empire of the Great Ming, was the ruling dynasty of China for 276 years (1368–1644) following the collapse of the Mongol-led Yuan dynasty
I think that covers China two empires collapsing about 50 years after the start of the LIA.
Americas
Maya historical accounts document two later great droughts between 1330-1334 AD and 1441-1461 AD. As a result of a long (documented) drought in Mexico and the Yucatan in the 16th Century (post-conquest), it has been estimated that half of the then Yucatan population died from famine and disease.
Long before Europeans made it to North America, the so-called Mississippians had build a great city surrounded by huge earthen pyramids and a Stonehenge-like structure made of wood to track the movements of the stars. Called Cahokia today, you can still see its remains in Illinois. At its height between 600-1400 AD.
South America: There are several theories as to what caused the demise of the Moche political structure. Some scholars have emphasised the role of environmental change. Studies of ice cores drilled from glaciers in the Andes reveal climatic events between 536 to 594 AD, possibly a super El Niño, that resulted in 30 years of intense rain and flooding followed by 30 years of drought, part of the aftermath of the climate changes of 535–536.[9] These weather events could have disrupted the Moche way of life and shattered their faith in their religion, which had promised stable weather through sacrifices.
As it happens I'm not a historian but I like to check things out for myself. It is clear from your replies your knowledge is very limited, you do very little research but parrot buzz pseudo science you've read on the Guardian website.
So unless you can come up with counter evidence I think I have more than answered your original question about warmer being better for the human race than colder (than present). Your models are just that, models.
Models superior to climate or economic models
One final suggestion, follow Entropic man's lead, we may disagree with him and find his school teacher background comes to the fore a bit but he always posts the references when asked (or has done when I ask up to now). We may then disagree as to whose reference is more likely to be correct and why but we have something concrete to work on.
Raff:
“Challenge models”? Wherever did you get that idea? Or is this yet another example of you moving the goalposts. NOBODY has “challenged” models; what has been challenged is the level of evidence that can be acquired from them. What I have mooted is that models can only provide evidence that the input into them is correct or not – and that can be vital information in many, many fields of science and research. Nobody has offered any argument against this – certainly not you; you do seem to be dodging any request for more constructive argument. What other forms of evidence could be obtained from models?
“The devil is in the details.” Very trite – and yet another example of you trying to hijack the argument, away from requests for you to provide us with this elusive evidence. Such details as blocking the construction of electricity generators in the poorer parts of the world; electricity that would help to provide cheap, reliable energy for the poorer people, and help them to drag themselves out of poverty – poverty, it would appear, that many of the so-called “Green” activists want to drive all of us into. Yes, the devil is very much in the details; SandyS has very helpfully provided yet more, and it would seem that there are many more details that, if you are not aware of, you certainly do seem to be ignoring.
Sandy, you have remarkable knowledge of historic events and it amazing how you can link these events to temperature variations. I had no idea that we could be so certain of global climate in the last 3500 years that we can make such inferences. Those tree ring and ice core studies must really have been perfected in the last few years.
It makes me wonder what else might have been temperature related. The US war of independence perhaps or the civil war? The rise of the British empire, perhaps, or its fall? The rise of the American empire? The rise of Islam or Christianity? The spanish civil war or the rise or fall of communism? It is exhilarating to think of all the possibilities.
What really amazes me though is that you can believe that variations of possibly as little as 0.5C globally might have been responsible for the rise or fall of empires or civilizations and yet be so unconcerned at the possibility of a 1 or 2C rise globally.
Radical, you don't like 'challenge'. Okay, I've reworded it for you:
It is odd that people who say models produce no evidence are also such keen users of model results as evidence to support their arguments. There was Rhoda with her certainty that cumulus clouds are negative feedback in action and that energy is distibuted faster (than something unspecified) by water in the atmosphere. Now Radical uses models against me. Warming is supposed by models to be concentrated towards the poles and therefore my argument that the effects of warming across the whole globe not just the colder areas need to be considered can be rejected. There is no warming in hotter areas he says (and by inference, there will be none) and so they can be ignored.
Yet if I look at spectral maps I don't see "no warming" in the hotter parts. And global warming is unlikely to leave regional weather systems of hotter parts of the planet unaffected. So you do need to consider what might happen in hotter, poorer places. They might get greener, but that doesn't in itself represent an economic gain - the devil is in the details.
Raff:
Your diatribe against SandyS is nothing but a straw-man argument, one that would be used by nobut a facile oaf. You asked for evidence that a warmer world would be more beneficial than a cooler one, and that evidence has been provided. Suck on it.
As for your insistence that, after attacking models, Rhoda and I use model results to back our argument: another straw-man, boy. If anything, it really shows what I have been saying time and time again – the ONLY evidence obtainable from a model is whether or not the suppositions in its construction were correct. I have asked you to provide an argument against that, and to enlighten us all as to what other evidence can be gained. As yet, you have done nothing to address those two basic, simple points.
You are, I suspect, carrying on with your utterly non-argument of taking pops at whatever others say without providing anything concrete yourself in the hope that we will get bored and stop replying. Thus, having the last word, you will have won the argument! Clever you! Something for you to proclaim loudly in the quad next October, perhaps.
Are you guys STILL wasting energy on this disruptive nonentity?
Using well defined PREVIOUSLY PROVEN parameters, the solar system can be modeled to show where a particular planet will be at some particular date in the future. Whether this result should be described as 'evidence' is perhaps a matter of semantics.
With climate being intrinsically chaotic in nature, and with many assumed and few well defined parameters, no climate model can reliably inform us of the temperature in x number of years. To adopt the semantics used in the thread, the climate model for a chaotic system, using poorly defined parameters, cannot provide 'evidence' of temperature at some specified future date.
Let's face it, there seems to be no 'consensus' on the actual value for climate sensitivity, which is of course one of the main parameters for climate models. We've been informed from 'settled science' that it could be up to 4C for a doubling of CO2, though this has been dropping lately as the pause has caused some embarrassment for the alarmists. The latest paper claims a sensitivity of 1C.
http://wattsupwiththat.com/2014/07/23/new-paper-finds-transient-climate-sensitivity-to-doubled-co2-levels-is-only-about-1c/
With such ill defined parameters as that climate models must come under the heading of GARBAGE IN - GARBAGE OUT.
Although this model looks interesting :-)
http://wattsupwiththat.com/2014/07/23/new-climate-model-introduced-now-with-knobs/
Raff,
actually I remember what I've read, and thanks to the internet , these days I can double check. Most of my post was cut and paste, I didn't say so to check your reaction; which failed to disappoint. Here are a couple more memory then internet trawls for you. You are quite right although I suspect you didn't mean to to be. Pre-industrial revolution society and civilisations were extremely vulnerable to changes in climate and environment some self inflicted (Easter Island). Using pre-industrial revolution technology to solve a non-problem produced by flaky models isn't going to help the human race when temperatures decline.
What really amazes me though is that you can believe that variations of possibly as little as 0.5C globally might have been responsible for the rise or fall of empires or civilizations and yet be so unconcerned at the possibility of a 1 or 2C rise globally.
The reason I'm not too concerned about a rising temperature is that globally the temperature is almost at a minimum, for the current inter-glacial, a drop of 0.5~1'C would have serious consequences a rise of 1~2'C would get us back to an average level.
Then there's George Santayana
Progress, far from consisting in change, depends on retentiveness. When change is absolute there remains no being to improve and no direction is set for possible improvement: and when experience is not retained, as among savages, infancy is perpetual. Those who cannot remember the past are condemned to repeat it.
and
Fanaticism consists in redoubling your efforts when you have forgotten your aim.
As you're an obvious sceptic then how about some UK Government data (no links you obviously don't read them)
An estimated 31,100 excess winter deaths occurred in England and Wales in 2012/13 – a 29% increase compared with the previous winter.
There were an estimated 24,000 excess winter deaths in England and Wales in 2011/12 – an 8 per cent reduction compared with the previous winter.
There were an estimated 25,700 excess winter deaths in England and Wales in 2010/11, virtually unchanged from the previous winter.
You can go back further in time but it is a similar story all the way down. This is a reflection of many things, quality of housing, income, energy and food costs being amongst them. There's no equivalent summer data, although what research I have read, in the UK excess winter deaths are several years early and deaths during summer heatwaves mainly affect vulnerable people already in ill health bringing forward the time of death a few months. I'm quite happy for you to prove that wrong with some linked information.
Martin A, you moved the goalposts. You suggest I or others think that "a model, ... can provide the same sort of evidence about the physical world that physical measurements and observations provide". Maybe you have somewhere, but I have not seen that said. And I have only said they provide evidence, unqualified.
on't move it.
Jul 22, 2014 at 11:14 AM | Unregistered CommenterRaff
Well, I don't think I said that exactly. What I said was: I am *guessing* that Raff would say that a thought coming into my head "I think it's going to rain this afternoon" provides "evidence" about rain, of the same general character as a measure of the volume of water being caught per hour in a measuring dish.
If you say so, then my guess about what you would say was wrong.
In that case, would you like to clarify how 'unqualified evidence' in the sense you use it, differs in general character from physical evidence from measurements or observation? A suggestion that something *might* happen would count as evidence? (I'm guessing again.) Or information that something will happen, unless something else happens in the meantime to prevent it, is evidence?
I'm trying to understand what you are on about but finding it hard to make headway.
[If I did not say a thing, you should not put it in quotes and attribute it to me]