Tuesday, February 03, 2009

Climate change is data-gathering

One of the things that yanks your humble Devil's pointy little beard is when those talking about climate change, anthropogenic or otherwise, keep burbling on about how they studied this science or that science and if you are not a scientist then you have no right to discuss the subject whatsoever.

As I attempted to intimate in this post, the only science at the heart of the climate change debate is that of data-gathering.

And, though the scientific community will no doubt be shocked by this assertion, you do not actually need shedloads of scientific training to be able to design and set up a reliable data-gathering exercise. Yes, some scientific training can be of use, but only because a stable foundation of said training is—or certainly used to be—practical experiment design.

Equally, you do not need to be a scientist to spot when a data-gathering exercise is fundamentally flawed. Let us take, for instance, the whole "global temperature" idea, shall we?

How does one measure global temperature? It's a bit tricky, wouldn't you say? After all, the Earth does not have one temperature across its entirety at any one time. As such, you need to measure the temperature in a particular spot over a reasonably long period in order to be able to gain an average temperature for that spot.

Given that the Earth has been in existence for about 4.5 billion years, you then need to ask yourself how you define "a reasonably long time"? NASA's Goddard Institute of Space Studies (GISS)—James Hansen's crew—uses a network of land temperature stations, some of which have been in place (roughly speaking) since 1880. That is to say, those stations have been recording measurements for just 128 years out of 4.5 billion, or 0.00000284% of the time that Earth has been in existence.

Of course, one could say that we are really interested in how temperatures react to man's activities, and vaguely civilised man at that. So, let's say that we have records of civilisation going back about 2,500 years. OK, so, these stations have been around for about 5.12% of the time that we want to examine. Is that long enough? Who knows?—it will just have to do.

So, having determined that 128 years is long enough, we need to set these stations up all over the surface of the globe, in order to get representative results. Well, the Earth has a surface area of some 510,072,000 km²: what density of measurement stations would be representative? And remember that you have an added problem: only 29.2% of that is land—the other 361,132,000 km² is water.

But you can't simply measure the surface: we are looking for a global temperature, so you need to place stations in the atmosphere too. Now, the Earth has a mean radius of 6,371.0 km and the volume of a sphere is calculated thus: v = (4/3) x pi x radius3. This gives us a rough volume of 1.3333 x 3.14159 x 258,596,602,811 = 1,083,205,999,191.998 km3.

But we want the volume, not of the whole sphere, but of the atmosphere that we can measure. So, we'll be generous and use only the three quarters of the atmosphere that is underneath the troposphere; the troposphere is at an average of roughly 12 km above the surface of the Earth. So, the volume, including the troposphere, is about 1.3333 x 3.14159 x 260,060,583,887 = 1,089,338,303,588.998 km3.

Now, to get the volume that we need to install our temperature stations in, we simply take the first total from the second: 1,089,338,303,588.068 - 1,083,205,999,191.998 = 6,132,304,396.07 km3. Which is a pretty large volume to measure, is it not?

Plus, of course, we haven't actually taken account of the depth of the seas in that calculation—after all, the Pacific Ocean alone adds another 161,000,000 cubic miles to the total).

NASA GISS relies on only a few thousand surface-only temperature stations, the oldest of which has been established for less than 130 years (and all of these are in the USA). Furthermore, as the now 67% complete surfacestations.org project has shown, the measurements from many of these stations have been contaminated by the spread of urban heat islands and other localised man-made factors.

And we haven't even delved into the problems of advancing technology (are thermometers more sensitive now than they were in 1880. Yep), incomplete records, human error and other such issues yet.

But in spite of all of this, the IPCC—which relies heavily on the NASA GISS data (though there are other sources too)—believes that it can tell us what an average global temperature should be. They cannot. Using the NASA GISS resources, they might be able to tell us what the temperature for one particular spot was, but only to a maximum of 128 years ago.

What to do? Well, our earth is old and so are many of the beings on it and we can try to use some of these as proxies to measure temperatures further back than our instruments or records can go. But these provide their own problems, as Michael Mann (amongst others) found with bristle-cone pines. And, Al Gore, famously and inconveniently, found that his proxy—ice cores—contradicted his thesis that CO2 caused temperature rises when it was discovered that the data showed that a CO2 rise lagged behind a temperature rise by an average of 800 years.

Anyway, in trying to illustrate the problems inherent in the climate change data-gathering exercise, I have turned this post into something of an essay (when it was meant to be a swift point) so maybe I should summarise: we don't really know what temperatures global temperatures are now, let alone what they have been to a reasonable past timescale—and we certainly do not know to the fractions of a degree Centigrade that are required for determining the current supposed rate of warming. We simply do not have enough reliable data to decide whether or not there is anything to explain, let alone postulate an explanation.

So, lengthy though this post has become, the point that I was trying to make is a pretty simple one: if we manage, somehow, to get some data that shows a trend that we need to explain, then we can start calling on the scientists who can then come up with hypotheses and properly test them.

But until we have some reliable data, there is no point in wittering on about your PhD.

50 comments:

Frank O'Dwyer said...

This whole essay is vigorous flailing at a strawman. Nobody said measuring the temperature was easy or that the data was perfect. The idea that we don't know everything therefore we know nothing is wrong.

The fact remains that all the data that does exist, including basic physics and many proxies, points to a warming trend. I'm sure it's all just a massive coincidence/conspiracy, but it seems to have fooled the ice into melting and much of the animal and plant kingdom seem to have been taken in by it too. If not, they are doing very convincing impressions of moving north to warmer climes, holding spring earlier, and going extinct.

According to new research, somehow the tropics have also been convinced to expand too - they mustn't read the intertubes, because they don't seem to be aware that the warming stopped 10 years ago.

"After all, the Earth does not have one temperature across its entirety at any one time."

Neither does the bulb in a thermometer. That must mean it's not really cold outside today, or at least we have no way of telling. Somehow we overcame this obstacle.

"ice cores—contradicted his thesis that CO2 caused temperature rises when it was discovered that the data showed that a CO2 rise lagged behind a temperature rise by an average of 800 years."

Just as the fact that chickens lay eggs contradicts the assertion that chickens come from eggs.

And since an engine is started by a charged battery, that contradicts the thesis that the engine charges the battery.

Devil's Kitchen said...

Frank,

"I'm sure it's all just a massive coincidence/conspiracy, but it seems to have fooled the ice into melting..."

When the Arctic sea ice gets the message that it should be melting consistently, rather than in a pre-established 30 year cycle, you let me know, eh?

In the meantime, you might like to look at the actual data: you'll note the down-trend, but the strong up-trend in the last few years.

Oh, and you do know that the Antarctic ice is at record levels, don't you?

"Just as the fact that chickens lay eggs contradicts the assertion that chickens come from eggs.

And since an engine is started by a charged battery, that contradicts the thesis that the engine charges the battery."


What? We are talking about an appreciable gap between said rises too, you know. These trends are absolutely nothing to do with chickens, eggs, motors or batteries. I may have set a straw man, but you're clutching at straws.

"The fact remains that all the data that does exist, including basic physics..."

Um, Frank? How does "basic physics" show that we have a warming trend? Basic physics might supply an explanation, but data? No. Please explain the method whereby "basic physics" supplies such data.

"... and many proxies, points to a warming trend."

Name one proxy (that has not been shown to have a divergence problem), and please set a link to the relevant paper.

Cheers,

DK


P.S. It seems probable that there has been a slight warming. Satellite data, certainly, seems to indicate it but they have only been monitoring reliably since '79.

The fact that there is a warming trend does not, unfortunately for AGW enthusiasts, prove that CO2 is the driver of said warming.

Further, as far as proxies tell us anything, we know that warming has happened before mankind was even close to evolving.

Thatcher-right MA (cantab) said...

One major thing you do get out of a scientific education is an appreciation of the scientific method: Observation – Hypothesis – Prediction – Test Prediction.

I have yet to hear one of the global warming mongers stand up with a prediction and say “if this doesn’t happen we have to accept that the whole idea is rubbish”.

I’ve heard plenty of scare stories “Sea level to rise by 350 m by the end of next week” (to exaggerate slightly) and plenty of retrospective justification “Obviously Hurricane Katrina was caused by Global Warming” but no predictions.

Indeed, there seems to be a move to back away from the name “Global warming” at the moment. I assume that this is because there is inherently a testable prediction in it!

p.s. one big advantage is having a Degree in Physics from Cambridge is that can win most scientific “mine is bigger than yours” squabbles!

Neal Asher said...

Sure, it is about data gathering, but it is also about how that data is sorted and applied. When Hansen and Mann do it they seem to apply a calculation called 'ideological forcings'. MacIntyre keeps catching these buggers out, but they just quietly correct the data and start setting up the next big lie to be broadcast to the believers in the media.

Frank O'Dwyer said...

DK,

"When the Arctic sea ice gets the message that it should be melting consistently, rather than in a pre-established 30 year cycle, you let me know, eh?"

I'm not even sure what standard you could mean by 'melting consistently' that the arctic ice trend wouldn't meet. Artic ice has been in clear decline for decades (even discounting the recent record, which may simply have been a freak event), as has global ice. Ice volume loss is particularly striking.

"the strong up-trend in the last few years."

A few years isn't a trend (and it is only a few weeks ago that the trend was below the record trendline for the same time). Weather is just weather. All the records (including temp) show short term noise that is different from the long term central trend. That would also be true in any such data that contains noise. It is a mathematical property of such data and there is nothing surprising about it, nor is it evidence of any long term change (not this data, anyhow, since the trend is not statistically significant).

"How does "basic physics" show that we have a warming trend?"

True it does not show it in the sense that an observation does, however it predicts it. Even if you had no thermometer at all you'd expect an increase in GHGs to result in warming, in the same way that you can figure out that you'll be warmer with your coat on without having to attach probes to your body. But of course in this case we also have the equivalent of the probes.

"We are talking about an appreciable gap between said rises too, you know."

You said the lag contradicts the 'thesis' of causation. It doesn't in general (hence my examples) and it doesn't in this case either. (You also said it was Al Gore's thesis, which of course is a nonsensical claim.)

Not only that, but it was the clueless climate scientists who predicted the CO2 lag before it was observed. Not bad for a bunch of people who apparently constantly make so many elementary errors that any amateur can spot in their spare time.

Indeed for a theory so flimsy that mere hobbyists can find holes in it with scarcely a thought, it is amazing that none of them have taken the time to get a paper published that refutes it, or managed to convince most scientists of it.

"Oh, and you do know that the Antarctic ice is at record levels, don't you?"

Yes, the antarctic is an exception and it is also a lot more complex. However does not provide much reason to conclude that warming has not occurred globally. It doesn't even provide a reason to conclude that global ice hasn't been in decline recently.

"Further, as far as proxies tell us anything, we know that warming has happened before mankind was even close to evolving."

Well, so did the big bang. I think my insurance premiums would be affected if we returned to those conditions now though, and if we were doing something that made that likely I would urge us to stop. Similarly with AGW. I prefer the conditions that we actually evolved to cope with.

Dave said...

The whole ecobollox joke is junk science.
Hanson's an egomaniacal idiot.
Gore's only in it for the money. He travels in a Lear jet for god's sake.

They are trading on the sheeple's ignorance and fear.
O'bama's bought the lie.

They can't just keep changing the facts to fit their madcap theories.

Any doubters please check out Wattsupwiththat or junkscience

And use your brains!!

Anonymous said...

I would like to make it clear that the current snow event did not originate in Britain but was caused by weather systems that are outside the government's control. Of course we are doing everything we can to help families and businesses weather the severe conditions and we will continue to take the right long-term decisions to deal with the snow. But I should like to point out that the snow is not as bad here in Britain as it is in other countries and we are well placed for when the snow thaws.

Thank you.

Gordon Brown

Tim said...

Does every school child not know that the Romans grew grapes in York 2000 years ago, because it was distinctly warmer back then?

So why on earth do people keep falling for this 'man-made CO2 makes the earth warm up' bollocks?

Utter mystery.

Gareth said...

It is simple hypocracy.

Be a skeptic and not being a scientist goes against you. Be pro-AGW and your qualifications or lack thereof is immaterial.

Be a skeptic and your source of funding goes against you. Be pro-AGW and being funded at the whim of politicians is immaterial.

'Scientists' have said so many different things in the last three decades. The data records are too short and too patchy, the computer models too crude.

Has anyone, anywhere done empirical tests to prove and quantify the CO2 effect on temperatures? Presumably they haven't (despite it being a relatively simple proposition) or it would be used to beat skeptics about the head with at every opportunity. Has anyone, anywhere put their hand in their pocket (or even ours) to spend a relatively piddling little sum to both improve the quality and scope of the surface stations? Many are clearly badly sited and in two cases (Canada and the former USSR) they have seen a substantial reduction in the number of sites being used. To put it simply - the fewer sites the poorer the data. Garbage in = Garbage out. Or if you're Jim Hansen, Garbage in = Warming out.

What we have here is science poisoned by politics. Scientists used to be happy to say 'We don't know enough' or 'We don't know everything'. Politics demands definitive answers and certain high profile activist scientists are happy to allege they have them.

Encourage energy efficiency - it saves people money. Encourage less waste - it saves people money. But in terms of scientific endeavours it is counter productive to both say the science is settled (when has it ever been thus in any scientific field?) and try to close down any opposing views. What are they afraid of?

saucepn said...

"Does every school child not know that the Romans grew grapes in York 2000 years ago, because it was distinctly warmer back then?"

Not under NuLieber Educashion they wouldn't.

Anonymous said...

er , 3 A levels and 3 degrees - all core science and I can't say I really understand all this global warming stuff. There mifgt be a teensy weensy bit of something in it but I am not convinced even of that.

Seems more of a quasi religious belief than science to me.
Try arguing with one and you get that "talking with a Mormon" feeling pretty soon.

Sargon the Demented said...

DK,

You mention surface stations, the paucity thereof; what you may not know is the number and quality of stations is declining rapidly.

Check this page out:

http://www.climateaudit.org/?p=2711

Cheers!
Ade.

Blue Eyes said...

When I presented my Masters report, my supervisor expressed surprise that we hadn't unequivocally "proved" the theory as most previous students had. He was a bit shocked when we told him we hadn't "cleaned up" the data...

I have come across sufficient PhDs to know that they are not superior beings - on average...

Anonymous said...

Listen to me yoiu sceptics Join Me and my friends at our next Blind black lesbians against global warming meeting and We can prove beyond any doubt the Earth is warming. Afterwards enjoy coffee and biscuits and our monthly challenge to turn Belinda away from her lesbianic ways.

Thatcher's Child said...

It is all about the difference between basic maths, a religious zeal, a love of funding and a Do as I say, not as I do mentality!

I get involved with government funding applications on a regular basis and the number of times the term climate change turns up in a grant application for anything is proportional to the final payout of the funding!

There seems to be an understanding in the current political class that there are somethings you must agree with publicly to ensure that you get to do what you really want to do.

The funny part of this is that in the outside world, there are mentalists like Frank who actually believe this nonsense, there are ranters like DK who think that the political class doesn't know that its all nonsense, and there are people like me who know that what either of you say is irrelevant because the only voice anyone listens to is the one with the money!

I understand the next big funding party will involve Eugenics!

Frank O'Dwyer said...

TC,

"I get involved with government funding applications on a regular basis and the number of times the term climate change turns up in a grant application for anything is proportional to the final payout of the funding"

a) Which is as it should be if it is really happening, as all the evidence indicates it is, therefore an entirely circular argument

b) It's not even true since the government in America has been actively suppressing climate science and indeed science of all kinds.

"mentalists like Frank who actually believe this nonsense...ranters like DK"

And then there are anonymous conspiracy nuts like yourself who use namecalling as a substitute for rational argument. At least DK argues his case, such as it is.

cookie said...

Frank said:
'Nobody said measuring the temperature was easy or that the data was perfect. The idea that we don't know everything therefore we know nothing is wrong.'

Hmm. That looks to me like putting words in the Devil's mouth. I'm pretty sure he didn't write or imply in his writing that somebody said that taking the global temperature was easy and that the data was perfect. My understanding is that DK is pointing out that the quality of the ground-based temperature data relied upon by some climate scientists isn't that great. A fairly uncontentious claim, I would say.

I enjoyed this part:
'And, though the scientific community will no doubt be shocked by this assertion, you do not actually need shedloads of scientific training to be able to design and set up a reliable data-gathering exercise.'

Heh.

Anonymous said...

AGW is finished, the recession killed it. Move on.

Frank O'Dwyer said...

cookie,

"My understanding is that DK is pointing out that the quality of the ground-based temperature data relied upon by some climate scientists isn't that great. A fairly uncontentious claim, I would say."

Since it is corroborated pretty well by the other sources, it would seem that it's not that bad either. Also, it's worth remembering that they are not trying to measure global temperature but the trend in it - there is a difference.

It's also a well demonstrated fact that inaccuracies and noise in measurements can be reduced by taking more of them. For example if you wanted to measure the length of a room with more precision, you'd get a better result by measuring it multiple times then taking the average.

cookie said...

Thanks for the lesson in minimising error, Frank ;) A minor point of correction - repeat measurements do indeed lead to greater accuracy (i.e. close to the true value) through the averaging of measurement errors; however, if you want a more precise measurement then you need a better instrument. A metre rule will never give you measurements more precise than to the nearest mm, no matter how many repeats you do.

Thatcher's Child said...

Frank said:

And then there are anonymous conspiracy nuts like yourself who use namecalling as a substitute for rational argument.

Well, In my world, everyone on the internet is a Dog!

Firstly, I wasn't making an argument, it was an observation.
When I was a child (30 odd years ago) I made an observation which all the AGW fanatics like Frank still don't understand - When I went to the big city from my home in the mountains, I was always surprised how much warmer it was, even though it was only 60-70 miles from home. I quickly worked out that it was all the heaters and the like which had made my coat a weighty irrelevance - until I got home again.

This observation is called the urban heat effect and is something that you can see when you take readings from a building which slowly get taken over by urbanization. Obviously, it takes years for this to happen - but it does.

The problem for people like Frank is that this toxification of the data is either missed, ignored, or encouraged, depending on the use of the data.

Data by itself is just random numbers with the pattern of a number of monkeys playing with calculators. If you play with them enough, without knowing exactly what you are measuring, you end up with a hockey stick - and that really would be a bad idea!

As Stewart Lee said - You can prove anything with facts!

Anonymous said...

‘corroborated pretty well by the other sources’

Actually, the GISS surface data seems to be diverging from the UAH satellite data, as is data from NOAA.

See here

Anonymous said...

I still can’t get over ‘the antarctic is an exception’.

We are about to commit $45 trillion to fixing a problem whose existence is based on the flimsiest of evidence, and he waves away what’s going on in entire continent because it doesn’t fit the ideology.

Frank O'Dwyer said...

cookie,

"if you want a more precise measurement then you need a better instrument. A metre rule will never give you measurements more precise than to the nearest mm, no matter how many repeats you do."

Not sure that's right, but it's not very relevant anyway, since the individual sensors are pretty precise - it is the noise that is large. And that can be (and is) reduced by more and (since it is the trend that is sought) repeated measurements and exclusion of outliers etc.

TC,

"This observation is called the urban heat effect"

You're not the only one who's heard of it. Even I've heard of it. This has been known to scientists for something like 200 years - and it's been thoroughly investigated and it doesn't make a shit of difference to the conclusion that it's warming.

"The problem for people like Frank is that this toxification of the data is either missed, ignored, or encouraged, depending on the use of the data. "

The problem for people like you is that it is not. There are many papers on this topic.

"Firstly, I wasn't making an argument, it was an observation. "

That you weren't making an argument is your only accurate observation so far.

Anon,

"Actually, the GISS surface data seems to be diverging from the UAH satellite data, as is data from NOAA."

So UAH is different to GISS+NOAA and you conclude GISS is the problem? Hansen derangement syndrome. There do appear to be problems in UAH. Wonder who runs that?

Anonymous said...

"Since it is corroborated pretty well by the other sources, it would seem that it's not that bad either."

The surface record is garbage, for numerous reasons ranging from poor measurements, poor placement of instruments, replacement of instruments by new ones in different locations, through to Soviet under-reporting of temperature in order to get more fuel supplies and over-reporting of temperatures near beaches to increase tourism revenue.

These stations were never, ever intended to be used to measure temperatures on a global scale, and I'd be surprised if you could reduce the average temperature station's record to an error of less than +/- 2C over the last hundred years; absent a time machine there is simply no way to determine how accurate the old measurements were or even whether someone just made them up for their own ends. The only reliable global record we have is from the satellites, and that only goes back to 1979.

"It's also a well demonstrated fact that inaccuracies and noise in measurements can be reduced by taking more of them."

Sure, you just need to back in time with your thermometer and take a few billion more measurements. Good luck with that one.

Henry Crun said...

@ Frank O'Dwyer

A couple of questions.

1. What is the average daytime summer temperature at the Antarctic?

2. Now that you know the answer to Q1, how much will the temperature at the South Pole have to rise in order for the ice to start melting in significant quantities to make any difference in sea level?

Dave said...

Dear Mr Devil,

I thought that your ambition for this world was to make it a much hotter place for everyone.

So how come you're not on the side of the doom-mongers?

Rush-is-Right said...

volume of a sphere is calculated thus: v = (4/3) x pi x 3(radius)

DK there's a mis-print (or typo) in this. It should read v=4/3 x pi x r cubed. You've clearly used the right formula in your workings but it doesn't look right.

Devil's Kitchen said...

Rush-Is-Right,

Cheers: I've corrected that. The site that I viewed when I looked up the formula did not do superscript, so they had written it as "r3".

I realised that the calculation was not correct, and thus cubed it when I was doing the figures, but forgot to change the written equation.

DK

weather it is nobler in the mind... said...

I am so cold I am shivering as I pull my fingerless gloves on, but the beads of sweat from global warming run down my face and drip on to my khaki shorts.

help!

Anonymous said...

Because the atmosphere is very thin relative to the globe, you can get a reasonable approximation of its volume by multiplying the earth’s surface area, 4 x pi x r squared, by the atmosphere’s thickness. The sum is bit easier to do and gives the same answer to two significant figures. It is what might be described as a Flat Earth Approximation.

6 billion cubic kilometres. The atmosphere is quite big, is it not? That’s about a cubic kilometre for each person.

Rush-is-Right said...

By the way, if you want to see what REAL global warming looks like, try this.

Rush-is-Right said...

And while I'm about it, this is interesting too.

Frank O'Dwyer said...

Anon,

"The only reliable global record we have is from the satellites, and that only goes back to 1979."

And yet the trend in that record agrees pretty well with the surface record you say is not accurate. Perhaps the problems you cite do not matter as much as you think they do.

(You do know it is the trend that is of interest, right?)

Henry,

"Now that you know the answer to Q1, how much will the temperature at the South Pole have to rise in order for the ice to start melting in significant quantities to make any difference in sea level?"

Your Q1 is largely irrelevant to your Q2, which in turn is largely irrelevant to the question of whether there will be significant SLR - e.g. 2C rise would be enough to eventually melt Greenland which would make quite a difference in sea level regardless of the fate of the Antarctic. That, if it went at all might require 4C or more of rise (notice it is surrounded by wet stuff and a significant part of it is anchored below sea level).

Nor do I know why you address either question to me since I made no claim about the antarctic melting.

Anonymous said...

There’s a pattern emerging here.

The GISS, UAH and NOAA data are observed to diverge, not corroborate. None of the expert researchers can explain exactly what is going on.

But Frank knows that the ‘problem’ is in the UAH data.


The urban heat island effect has clearly affected a significant proportion of the USHCN temperature data. Even Hansen’s team makes adjustments to try and compensate for it.

But Frank knows that it ‘doesn't make a shit of difference’.


The Arctic was warming, while the Antarctic overall stayed about the same. None of the scientists can explain why this happened.

But Frank knows that it is the Antarctic that is an ‘exception’.


It is truly like debating with a Jehovah’s Witness.

Anonymous said...

Thatcher-right: spot-on about the benefits of a scientific education.

And I think we have done the following experiment:

We continued to increase the Carbon Dioxide in the atmosphere (slightly, and slowly, it's true, but increase nonetheless). And we watched to see what would happen to the temperature.

And what did we find?

It stayed the same for about ten years (roughly, from 1998 to 2008), and then it began to decline.

So that's the AGW hypothesis blown out of the water then.

What, I wonder, will be their next pretext for taking all our money and telling us how to live our lives?

Frank O'Dwyer said...

Anon,

"The GISS, UAH and NOAA data are observed to diverge, not corroborate. But Frank knows that the ‘problem’ is in the UAH data."

Try not to lie about easily checked facts. According to the commenter I was responding to (which may have been you): Actually, the GISS surface data seems to be diverging from the UAH satellite data, as is data from NOAA. In other words, that commenter stated that UAH was the odd one out. No doubt he concludes the problem is GISS, even though UAH is the one with the history of a sign error, and the long term trends in all of them agree pretty well.

Plus the divergence is not very significant if you're looking at the 30 year trend, which you would be if you were a climatologist and not an anonymous blogger incapable of reading for comprehension.

"The urban heat island effect has clearly affected a significant proportion of the USHCN temperature data. Even Hansen’s team makes adjustments to try and compensate for it.

But Frank knows that it ‘doesn't make a shit of difference’."

Anybody who's looked into it knows that. Why don't you? There are several papers demonstrating that it doesn't make any difference to the conclusion: it's got warmer.

Studies that have looked at hemispheric and global scales conclude that any urban-related trend is an order of magnitude smaller than decadal and longer time-scale trends evident in the series (e.g., Jones et al., 1990; Peterson et al., 1999). This result could partly be attributed to the omission from the gridded data set of a small number of sites (<1%) with clear urban-related warming trends. In a worldwide set of about 270 stations, Parker (2004, 2006) noted that warming trends in night minimum temperatures over the period 1950 to 2000 were not enhanced on calm nights, which would be the time most likely to be affected by urban warming. Thus, the global land warming trend discussed is very unlikely to be influenced significantly by increasing urbanisation (Parker, 2006). ... Accordingly, this assessment adds the same level of urban warming uncertainty as in the TAR: 0.006°C per decade since 1900 for land, and 0.002°C per decade since 1900 for blended land with ocean, as ocean UHI is zero.

- IPCC FAR

"The Arctic was warming, while the Antarctic overall stayed about the same. None of the scientists can explain why this happened.

But Frank knows that it is the Antarctic that is an ‘exception’. "

Of course it's an exception. What do you think the word exception means? Read what I wrote, not what you think I wrote.

Frank O'Dwyer said...

Anon,

"We continued to increase the Carbon Dioxide in the atmosphere (slightly, and slowly, it's true, but increase nonetheless). And we watched to see what would happen to the temperature.

And what did we find?"

Who cares what YOU found. There's little evidence you even understand the question.

People who know what they're talking about found that it's been steadily warming since about 1975.

Pogo said...

Frank... A quick question. Why has the concept of a "30-year trend" become the touchstone of the AGW community? Why not 50 years? Or a hundred?

It couldn't be because it's been cherry-picked as the best period that at present shows an ascending trend?

Pogo said...

Picking up on the IPCC quotes on UHI...

Peterson, Jones and several others appear to suggest that UHI is irrelevant, yet, for example, Some effects of the urban structure on heat mortality (1972), John F. Clarke of The Division of Meteorology, National Environmental Research Center, Environmental Protection Agency, indicates that night-time UHI may be in excess of 4C.

Interestingly, having read a few of the IPCC UHI cites it appears that one thing that's lacking is a bit of good old-fashioned experimentation - the papers seem to base their results on examining night-time satellite pictures in order to decide what is urban and what is not and then recomputing "homogenised" results from weather stations in the various regions. This, bearing in mind the frighteningly poor standard of much of the USA's ground station network, would tend to suggest that there's an element of averaging the result of "adding apples to oranges" going on. One would have thought that a few incidences of trolling a calibrated thermometer across an urban conurbation and noting the results might, just possibly answer the question once and for all. Problem is, satellites and supercomputers are "sexy", thermometers and bicycles aren't.

On reflection, IIRC some bloke who writes a weather blog (can't remember who) actually did something similar for one of his children's science projects and actually produced quite edifying results, but, as they weren't peer-reviewed (a procedure frequently the same as GCSE students being asked to mark their friends' papers) they're not really acceptable.

It would be interesting to see the experiment done properly under controlled conditions.

Frank O'Dwyer said...

Pogo,

I don't think there is anything set in stone about 30 years - it is a figure used by the WMO, and is a period long enough to eliminate year-to-year weather variations. There may be shorter periods that do the same (IPCC AR4 uses 20 at times).

But it's clear for example that 1-7 years is too short (as is 1 month or 1 week, as sometimes preferred by some on the 'sceptic' side of the aisle).

However since climate is basically 'average weather' I would think any time period would do as long as the trend is statistically significant for whatever part of the climate system it is you're talking about. For example if the temperature genuinely and significantly dropped like a stone for 3 years running, I don't think that could be dismissed as 'weather'.

Re UHI and this:

Peterson, Jones and several others appear to suggest that UHI is irrelevant, yet, for example, Some effects of the urban structure on heat mortality (1972), John F. Clarke of The Division of Meteorology, National Environmental Research Center, Environmental Protection Agency, indicates that night-time UHI may be in excess of 4C.

Irrelevant (to the trend) is not the same as doesn't exist. The daytime temperature in the oven in my kitchen is sometimes as much as 200C, and that is irrelevant to the warming trend too.

Anonymous said...

He stated 'the GISS surface data seems to be diverging from the UAH satellite data'. No one is sure why, or which one is 'right'. They could all be right, as they measure different things in different ways. But there is some divergence.

'omission from the gridded data set of a small number of sites (<1%) with clear urban-related warming trends' is prima facie evidence of at least some concern about urban warming. surfacestations’s survey, which post-dates those papers, looks likes it’s suggesting that 1% was not enough, as so far 69% of sites surveyed may be subject to >2C error. The question that can’t yet be answered, because the survey hasn’t been finalised, is when proper compensation is made for urban heat island effects, what is the underlying warming trend.

Half the world is warming. Half the world is not. Which half shall we call an ‘exception’? Refer to the dogma.

Pogo said...

The thing about trends is that they are altered by the length of the time period used... A 30-year trend might indicate an increase, whereas a 50-year might indicate a decrease and it's sometime a bit too easy to cherry-pick start and end points in order to make a case.

As to the UHI, I realsie that in theory it should be irrelevant to the trend, but as it's being adjusted for as part of the trend calculation it's better that the adjustment be in the right direction - and it would seem from Peterson's paper that not infrequently rural temperatures are being adjusted upwards to account for UHI - somewhat counter-intuitive methinks.

Frank O'Dwyer said...

Pogo,

"The thing about trends is that they are altered by the length of the time period used... A 30-year trend might indicate an increase, whereas a 50-year might indicate a decrease and it's sometime a bit too easy to cherry-pick start and end points in order to make a case."

Yes it might in principle (tho I doubt it does in fact). But then if you use purely mathematical techniques to let the data pick its own inflection points you can avoid cherry picking. Do that with the temp data, and you'll still come up with 1975 as a step change (e.g. see tamino's blog).

But there's no doubt at all that the short trends which are typically used by the 'sceptics' are blatant cherry picks...you can see that in how sensitive they are to adding even a single year's data, or how the error bars on the trend include 0, or how they always start from a record high point like 1998 to make the odds favor a decrease.

Another way of spotting such cherry picks is you'll usually see similar short trends repeated many times previously in the same data, as part of a longer trend which is clearly warming. Another example is DK pointing to the last few years of ice trend as an uptick, even though the record previously is full of exactly such, and the subsequent overall trend was still downward.

Last but not least, you can actually generate artificial data, for example using excel, which includes a trend + noise by construction. You'll easily find short 'cooling' trends in such data, even though in that case you *know* it represents warming.

"As to the UHI, I realsie that in theory it should be irrelevant to the trend, but as it's being adjusted for as part of the trend calculation it's better that the adjustment be in the right direction"

Which it must be (or at least any error cannot be very significant) since other sources lead to the same conclusion, as does ocean temp where there is obviously no UHI. Don't forget the question here is simply has it gotten warmer over the last 30 years or so, and leaving aside causes. There is so much evidence for this even without the surface record that it is silly to deny it. It's certainly silly to say that we have no clue whether it's warmed or cooled, or that UHI has not been considered at all when it clearly has. Not saying you have made these claims, but others here have.

Anonymous said...

UHI is only irrelevant to the trend if the effect is vanishingly small; or it is consistent over decades; or the error is randomly distributed; or proper compensation is made to eliminate it. Otherwise it introduces a systematic error. Far from being negligible, I’d guess it contributes at least several degrees increase in temp in some cases. We can easily see urban development occurring over the years, so we know there is no consistency. The bias is predominantly one way: an increase. It is virtually impossible to determine what the proper compensation should be, as it varies on a site-by-site and year-by-year basis depending on the extent of development, so a bias is almost impossible to avoid.

Satellite data is taken over a wider area and so should avoid some of the issues with urban placement of surface stations. It is single automated device (with two instruments measuring slightly different things (UAH and RSS) and not necessarily agreeing) so some classes of error are eliminated, but calibration errors and drift could still occur. No doubt, the sensor package designers have thought about that.

Of course some experimentation would help settle the UHI issue. But that seems about as likely as the Pope carrying out an experiment to prove there is no God. Eventually some enthusiasts will carry out the simple experiments you suggest; then they will be derided for being amateurs.

Henry Crun said...

Frank, time to give it up and get back to your deckchair on the beach and tell the tide to go out again.

Frank O'Dwyer said...

Henry, when you posed your questions, I thought you were probably being disingenuous and you had no interest in any answer.

Now I know.

Can you tell me where are the real sceptics? You know, the ones who bring the same scepticism to both sides of the argument, and actually change their minds when faced with evidence?

Henry Crun said...

Frank, does that mean you will chage YOUR mind when faced with the evidence theat man-made global warming is absolute cockwaffle?

Anonymous said...

I totally agree, Devil! It's posts like this that really highlight the rational basis of so much anti-global warming advocacy!

I find your post analogous to me making soup. When I have made some tomato soup, patiently cutting onions, sauting them in a knob of butter, adding quarters of ripe, juicy plum tomatoes and adding stock and cream, I often have a problem. Does my soup have enough salt in it?

Oh, sure I could taste it. I could dip a spoon in, but I would be only supping on less than a thousandth of the soups volume, or to be scientific with you > 0.1%. Although some scientists *eyeroll* would have you believe that the saltiness of my soup is a eminently measurable quality, I think we can both agree that 0.1% is a very small quantity. Things are much more complicated than that. Tasting a single spoon of it and drawing inferences is the very definition of anti-science.

I suppose a "scientist" would create a model whereby the spoon I taste is representative of the soup as a whole. This is just the kind of ridiculous stupidity I associate with scientists. I myself am a crystal hydropathist.

Anyway, I solve the problem the rational, holistic way. By making two identical pans of soup I can drink one of them in its entirety and thus be sure whether to add more salt to its twin. I have to go now as I feel rather sick.

Anonymous said...

Except that with your soup, crystal hydropathist *eyeroll*, you stir it first to make it homogenous, so a single sample is likely to be reasonably representative of the whole pan, at least to the extent that you can taste. Your teaspoonful is probably about 5ml in 500ml, or 1%. If you’re not sure, you might take several samples and average them intuitively.

If you make soup in a 5 ton vat, like some people do, you might want to take a bigger sip. You might even use an instrument to measure it. Of course, drinking the whole vat would be really difficult and wouldn’t leave anything to sell, but fortunately some smart mathematicians long ago worked out how to calculate the level of confidence that can be placed on a sample of any particular size. So you’d look up their work and use it to design a sampling plan that gives you the certainty you need.

For that, though, a crystal hydropathist might need the help of a scientist.

Unfortunately, we don’t have a way of making the whole atmosphere homogenous, otherwise a single temperature measurement would suffice. So we have to take thousands of measurements, each one representing perhaps one billionth of the total global atmosphere. What do we find? Every one is a bit different. What does it mean? That’s the question that the scientists are trying to answer.