Feeds

back to article Forget 'climate convert' Muller: Here's the real warming blockbuster

If new techniques endorsed by the World Meteorological Organisation are applied to official figures, over half of the global warming reported by US land-based thermometers between 1979 and 2008 simply disappears, researchers have found. The new study used the same raw temperature measurements as US government federal scientific …

COMMENTS

This topic is closed for new posts.

Page:

Anonymous Coward

In before "koch funded deniers working for oil industry deny THE SCIENCE and want to eat your babies!".

6
2
Silver badge
FAIL

hmm

Actually the timing of this study is perfect. It is really important to get out in front of the problem of one of the worst droughts in living memory in the US before crop failure causes food prices to double and people start asking uncomfortable questions. The key is to turn the public's attention towards something else. Nothing to see here move on.

0
0
Gold badge
Thumb Up

Data collection in the *field* and data reduction to the latest *international* standards

This sounds like proper science.

I'll point out however that just lopping out "low quality" sites would seem to be excessive. They should definitely have a lower weighting. How much lower requires more background knowledge than I have.

I'll also note that the field work gives *snapshot* picture of the sites. It's hypothetically possible that a sudden spurt of intense new development could have shifted a high quality site downscale in the recent past.

As always *context* is important.

Thumbs up for the solid effort. Weather it will be incorporated into the relevant models in this case may be more doubtful.

16
1

Re: Data collection in the *field* and data reduction to the latest *international* standards

I agree with your sentiment, though I now have more doubts about climate science than before. My doubts come not from a presupposition about the validity of AGW or not, but from a true appreciation of just how completely impossible it seems to ever properly account for the circumstances surrounding data collection or to control for the inherent anomalies that are bound to exist.

If the data can never be trusted, how can the science? Personally, I love science in general and appreciate the rigor that the scientific method and peer review maintain. But this seems like such a fool's errand. Better fun would be had jousting with windmills and dreaming the impossible dream.

5
2
Bronze badge
Headmaster

Re: Data collection in the *field* and data reduction to the latest *international* standards

"Weather it will be incorporated into ..."

I saw what you did there ;)

7
0

Re: Data collection in the *field* and data reduction to the latest *international* standards

"Weather it will be..."

Well played.

0
0
Silver badge

But there are too many low quality stations

Even given a small weighting, low quality stations are so numerous that they will still end up distorting the numbers.

It is unfortunate that there are so few high quality stations with a long history.

Airports have always recorded temperatures because temperature is an important factor in calculating take off weight. But unfortunately airports change with time.

SFO started off as a grass runway in an old cow pasture. Now it is a few square miles of concrete and asphalt with air conditioners belting out heat. No useful data in that.

5
0

Re: Data collection in the *field* and data reduction to the latest *international* standards

context seems to be everything.

They lop off all the sites they don't like, and end up with a number (notably, at least in this piece, without error bars. So somewhere in the 0.13-0.14 per decade with their picked numbers, purely for the US land)

What's that? Surprisingly low? Let's check:

http://www.reportingclimatescience.com/news-stories/article/global-warming-rate-is-014c-per-decade-says-uah-team.html

"On 25 November 2010 the UK Meteorological Office in a pre-Cancun press conference stated that it believed the increase in average global surface temperatures has been around 0.16C per decade between 1970 and 2000 but has ranged between 0.05C and 0.13C in the last ten years. The Met Office said that the trend figures for the period between 2000-2009 were based on data from three different source; one from NASA (0.13C), one from the US National Oceanic and Atmospheric Administration (0.07C) and one from HadCRUT3 (0.05C) - the dataset managed by the Met Office and by the University of East Anglia's Climate Research Unit"

So basically, a group of sceptics have looked at the data, and come out with something that still shows climate change, and in a range which is comparable to other studies, albeit non worldwide so it's apples to oranges.

But if you read the spin on this article, you'd think they'd proved that no warming was happening.

8
4

This post has been deleted by its author

Thumb Down

Re: Simple! Just ignore data you don't like

No. The problem is that the environment does not remain static. You might site a station in a field and 25 years later that station is in the middle of an airport or a large town, and is far hotter. The question is how to adjust the temperature to take this into account?

Surprisingly people like James Hansen have been doing this for a very long time. They mostly revised up recent temperatures and revised down past temperatures, to exaggerate the trend. What Watts et al has done is show using modern techniques that at least half of the warming trend is spurious.

21
3
Silver badge
Flame

Re: Simple! Just ignore data you don't like

...It's fine to try and disentangle your equipment from interference, but surely the whole point of a weather station is to capture the data as it is, and not as it might be wished?..

Ah. Yes, if you want to determine the actual temperature at a point. For instance, it's important for aeroplanes to know what temperature it really IS at Heathrow.

But if you want to know what the UNDERLYING temperature of the Earth SHOULD be, so you can see if a little CO2 is increasing it, you need to measure the temperature well away from any extraneous heat sources, like jet exhausts, air conditioners or big black tarmac car parks. Now, no one is going to pay to establish a set of thermometers in carefully controlled rural places ONLY, so we have to make do with the pre-existing local weather station. And aren't we lucky - that used to be a rural station 50 years ago, but now it's conveniently situated at the junction of two motorways. And, surprise, surprise, the average temperature has gone up a lot from 50 years ago.

Must be the CO2! Grant money, here we come! Closely followed by a lucrative post advising politicians how to tax people....

15
6
Anonymous Coward

Re: Simple! Just ignore data you don't like

You might also read the results of Watts' quite valid experiment regarding the affects of repainting Stephenson Screens (?) with modern paint rather than the former lime based whitewash. You guessed it, the temperatures rise. So every box that has been repainted since the 70's almost certainly exhibits a spurious warming trend.

I have always maintained the GW thing assumes warming and that the "warming" is very likely not particularly well documented. It seems more and more research confirms this.

If it's not warming, then CO2 can't be causing what is not happening ... a classic case of looking for the murderer before the body is in evidence.

7
2
Silver badge

Re: Simple! Just ignore data you don't like

"They mostly revised up recent temperatures and revised down past temperatures, to exaggerate the trend"

Claims like this are devoid of evidence.

The source code for Hansen's algorithm has been available to download for years (http://data.giss.nasa.gov/gistemp/sources/). If FORTRAN isn't your thing, someone has even ported it to python (http://clearclimatecode.org/)

Should be easy then to show everyone the part of the code that "revised up recent temperatures and revised down past temperatures, to exaggerate the trend". Yet no-one has ever demonstrated it...

12
3

Re: Simple! Just ignore data you don't like

Well, that's exactly what NOAA appears to do with its homogenization process: Identified the cool-running stations (that just so happen to be, on average, well sited) as outliers. And then adjusts their trends warmer.

What I did was rate the stations by quality -- regardless of their high or low trend -- deliberately concealing the trends from view when doing the ratings, in fact.

THEN I examined the trend to see if bad siting makes a difference.

Comprende, comrade?

11
1

Re: Simple! Just ignore data you don't like

if the weather station reports a high temperature because of ashpalt or not, is of secondary importance.

Because the issue is about the change in temperature over time, if they are high at the start of the period because of X environmental factor, they will be high at the end of the period also. The impact of this elevated reading is secondary because its the relative difference which is important here, not the absolute values.

1
6
Anonymous Coward

Re: Vidura

If you have a rural station that ends up as urban then there will be a spurious trend in the data. The relative difference will have changed due to urbanisation.

If you have an urban station that has an increase in population and/or buildings in the area then the relative difference will have changed due to increasing urbanisation.

The only way urbanisation is not an issue is if everything remains the same over then entire time period you are measuring.

6
1
Silver badge

Re: Simple! Just ignore data you don't like

"Because the issue is about the change in temperature over time, if they are high at the start of the period because of X environmental factor, they will be high at the end of the period also."

And what if environmental factor X wasn't present at the start of the period? For example, the weather station was situated in a field, but then development has occurred and it now sits in the middle of a housing estate?

2
0
Mushroom

Re: Simple! Just ignore that the data you like has been pre-adjusted.

"Should be easy then to show everyone the part of the code that "revised up recent temperatures and revised down past temperatures, to exaggerate the trend". Yet no-one has ever demonstrated it..."

It's not Hansen's algorithm that's being investigated in this paper, it's the data that's being fed to the algorithm. If that data has been pre-warmed, as Watts et al appear to show, then it's a simple case of GIGO. GISS's ultra-distance Kridging around the Arctic aside. It's a problem that is also not constrained to the US given poor station siting appears to be a global, man-made phenomena. It may also appear that blind faith in data quality is also a widespread phenomenon within mainstream climate science.

4
1

Re: Simple! Just ignore data you don't like

Yes.

Also, if the environment has NOT changed, is the trend (sic) higher for the poor stations. That is exactly what we are measuring.

It is.

0
0

Great job! Soon the climate hoax will be exposed. =)

Great job! Soon the climate hoax will be exposed, and then perhaps, we can start to worry about more important things such as our decreasing freedom, political power concentration, jobs for the young, and building thorium reactors for cheap energy.

15
17
Silver badge

Re: Great job! Soon the climate hoax will be exposed. =)

Climate change isn't a hoax, you'd have to be a fool to think our climate is static. Certain groups are manipulating the data and public opinion for their own gain though. I can't agree more about the thorium reactors, we need to stop messing about and modernise our energy supply.

14
0
Mushroom

Re: Great job! Soon the climate hoax will be exposed. =)

Climate change isn't a hoax, as climate is indeed changing and it seems always has and will.

Global warming isn't a hoax, as the world is indeed getting warmer and has been since the end of the last ice age.

But using "catastrophic anthropogenic climate change" as the basis for crippling the industrial economy of the developed world IS a hoax-based power and money grab that makes all previous such grabs seem puny in comparison.

15
7
FAIL

Re: Great job! Soon the climate hoax will be exposed. =)

@ fibbles 19h18z

But he didn't say climate CHANGE was a hoax - he said the climate hoax

Climate has always changed - its been hotter and colder than now; its the Catastrophic Man Made bit that is the hoax

We now have some access to both the original unadulterated data and also to a new datset that has traceable; repeatable documented changes made to it. Compare that with CRU & co's refusal to divulge either original data or what hocus pocus they got up to with it.

And as the IPCC will need reminding - empirical measurements trump models ever time. The empirical data (not models) indicates that the null hypothosis (current climate change is with in natural bounds) is valid. No need for fancy complex hypothesis that rely on bad models.

3
2

But isn't the fact

That we are becoming more & more urbanised a reason to include the data from these other weather stations in itself? After all its is "human behaviour" that is causing the temperature rise according to the climate scientists and greenies alike.I bet if you check the Heathrow airport weather station records against the rest of the UK you will find it represents the UK national average temperature pretty well.

2
13
Thumb Down

Not likely

The phrase is "urban heat island", and if you're not familiar with its meaning then you don't have all the information you need to form an opinion.

The effect on weather station readings is considerably greater than epsilon, too. For example, in the city neighborhood where I live, the temperature is generally three to five degrees hotter than it is in the tree-lined county suburb where I work.

If you're normalizing to one or the other, you have the choice of revising the city figure downward, or the county figure upward. It may seem as though either choice would be as valid, but areas of urbanization large enough to produce the heat island effect do not make up the majority of land area either in England or in the US -- in both countries, and everywhere else on Earth as well, the area not subject to the effect is orders of magnitude larger than that which is.

This being the case, Hansen's choice to revise upward figures from non-urbanized areas, rather than reducing figures from temperature stations in built-up areas, or producing a weighted average of the two based on land area, looks awfully questionable -- especially from the fellow who's been leading the charge in favor of the contention that "the science is in".

17
1
Anonymous Coward

Re: But isn't the fact

One of the problems with the surface temperature record is that oftentimes weather stations started up on rural grass aerodromes which then became concrete airfields surrounded by urban sprawl - e.g. what is now Heathrow was originally the "Great West Aerodrome" in the 1930s. Add to that the effect of e.g. air conditioning plant being put right next to the weather station and you can start to see why the recorded temperature might rise a lot more than what you'd see if the location stayed as it was, or moved a few fields away. Have a look at the surfacestations.org site and you'll see what I mean. So I'd expect a station at Heathrow to have shown greater warming than a similar site that wasn't subject to similar growth (but this is an informed guess, I haven't checked - do post the findings if you do so).

The Watts et. al. paper highlights how, in the USA at least, the process used to process the raw station data actually makes the warming look about twice what it actually is. The official adjusted data actually shows greater warming than the worst stations show before the adjustment. So Heathrow could even reflect the "average" - but the average might well not reflect the reality. The paper is definitely worth a read, certainly of the power point summary at least.

Incidentally, for a UK temperature record, have a look at the Central England temperature record on the UEA site. It basically shows two periods of warming in the last century with a bit of cooling in the middle. I've yet to see a convincing explanation of why the first period is different from the second, although it was before we started doing major "carbon" emissions.

10
0
Silver badge

Re: Not likely

"Hansen's choice to revise upward figures from non-urbanized areas, rather than reducing figures from temperature stations in built-up areas, or producing a weighted average of the two based on land area, looks awfully questionable"

I think you are completely wrong in your description of Hansen's algorithm.

See here where it is described, in particular this part:

"in step 2 they adjust the non-rural stations in such a way that their long-term trend of annual means matches that of the mean of the neighboring rural stations. Records from urban stations without nearby rural staitons are dropped."

http://cdiac.ornl.gov/trends/temp/hansen/hansen.html

Non-rural stations are adjusted to towards rural stations. When there are no rural stations the non-rural stations are excluded.

Can you explain where you got the impression that the opposite adjustment was being made?

7
1
Facepalm

Re: Not likely

Fair question. I got that impression from the caption of the first chart shown on Page 1 of Mr. Orlowski's article; specifically, "Instead of adjusting the poorly sited station trends downward to the levels of the well located stations, the well sited station trends are adjusted upward to match the poor station trends." This is the direct converse of the normalization algorithm described in the GISTEMP overview you cite.

That said, I was under the impression that that entire image came directly from NOAA, and on review I gather that's only true of the map itself; it appears that Mr. Orlowski, or someone else in the bowels of El Reg, added that caption. This being the case, I now find myself wondering who, if anyone, is misrepresenting what, from whom.

On a closer reading, I find that the paper under discussion appears to contend nothing about Hansen's algorithm, but rather that the data to which that algorithm is applied are of lower quality, due to inclusion of data from poorly sited stations, than those derived specifically from stations meeting the WMO standards for good placement. I appreciate you pointing out that my understanding of the discussion was erroneous.

I am still curious what might be the story with regard to that caption I mentioned, but I suspect the problem here is not that there's anything incorrect or misleading in the article, so much as that I should probably not expect, over the course of a couple of five-minute breaks from what I've been working on all day, to gain a solid understanding of a complex discussion like this one.

5
1
(Written by Reg staff) Silver badge

Re: Re: Not likely

"That said, I was under the impression that that entire image came directly from NOAA"

The heat maps come from NOAA, the in-graphic comment is from Watts - as the caption says.

C.

0
0
Silver badge

Re: Not likely

Ah ok its just a slight mixup then. Hansen is GISTEMP which is NASA. Hansen/GISTEMP is a different record than NOAA.

On the the otherhand I don't agree with the caption in the figure one of the article you have highlighted. It makes it sound like NOAA has deliberately adjusted well located stations in the wrong direction toward poorly sited stations. Unless NOAA had access to time travel they couldn't have known which stations this paper would classify as well located and poorly sited in order to do that.

Of course the claim could be that NOAA inadvertently adjusted incorrectly, but my impression is that isn't the angle the article is intending to portray.

0
0

Re: Not likely

Whoops. Thanks again -- that'll learn me to try to read and comment in five-minute bites of time, or should do, anyway.

0
0

Re: But isn't the fact [Facts and Factors]

But the factors affecting the stations are NOT representative of the topography the stations purport to represent.

10% of the stations we rated are urban and 25% semi-urban. That is an over-representation approaching 500%. And 6% of rated stations are ASOS (i.e., bad equipment) in airports. Rounded to the nearest percentage point, zero % of land surface is airport environment.

So, bad mesosite (regardless of microsite, i.e., Class 1 - 5 ratings) dominates fully 40% of the USHCN surface record.

3
0

Re: Not likely

Actually, I made the maps.

And rated all the stations.

And calculated the numbers in each region on the map.

And gridded the data.

(And wrote the comment in question, for that matter.)

NOAA has no idea what the trends are for Class 1\2 vs. Class 3\4\5 using Leroy (2010) methodology. They never bothered to find out.

9
0
Anonymous Coward

Re: Not likely

"Actually, I made the maps"

and so I see you did - well done and thank you Sir :-)

2
0
Bronze badge

Re: But isn't the fact

@GreggS

According to this:-

http://www.bbc.co.uk/news/uk-18623096

only 2.27% of England is actually built on. For the whole UK it is less than 1%. So, it would seem that using only records from urban areas might not be the best way to see the overall trend.

5
0

Re: But isn't the fact

I agree, but my point was that mankind(or womankind) is increasing its footprint and isn't the point of this after all how it affects mankind(or womankind)? If the majority of mankind(or womankind) live in Urban areas and those urban areas are showing an increase in heat, then it's a self fulfilling propechy that mankind(or womankind) is indeed warming the Earth.

0
3
Facepalm

Re: But isn't the fact

Not the point. The issue is stability of the readings over time, not how representative they are today. It could be that LHR weather is a good average for the UK (I have no idea), but when trying to see very slow trends it is important that everything except the measurement remains as constant as possible, or at least changes in surroundings, equipment etc. are also recorded and corrected for - some corrections will be done by regular calibration, others can't be calibrated out.

Regular updating of the metadata may make the low quality data usable over time, resulting in better results from a wider base.

Doh because this should have been started years ago, measuring a small number of variables when analysing a complex system is not good science.

2
0

Still warming, but less so

Looks like the warming trend with the more accurate weather station data still shows an increase, but only about half what's been claimed.

Which is fair enough, but one does feel moved to wonder, considering that "the science" is supposedly "in", why it takes an independent analyst to point out what we're looking at here, rather than us hearing it from the official organs tasked with telling governments what to do about climate change. The figures appear to be pretty unambiguous, and the data we're looking at here does appear to support the contention that the climate is changing (though anthropogenicity remains still very questionable), so why haven't we heard about this before now?

11
0
Bronze badge

Re: Still warming, but less so

Well, there's been little official help for Watts and the volunteers of surfacestations.org. Indeed, I recall that, when Watts originally announced the project in 2007 (yes, really, this isn't a recent thing) the station database was removed from public access within a couple of weeks, and Watts had to resort to legal action to regain access.

The trouble is that the 'establishment' don't want this made public because it will dilute the message and the fact that it comes from Watts, the denier-in-chief adds insult to injury. I expect that the 'official' response will be that, yes there are problems with some stations in the USA, but:

1. It's only in the USA and there are plenty of others worldwide.

2. We're in the process of fixing those in the USA

3. We've carefully calculated adjustments to account for station location. The procedures used by the independent BEST team confirm their accuracy.

4. Other peer reviewed papers have shown that UHI is negligible and can be ignored.

5. There's plenty of other peer reviewed evidence that AGW is a real and present danger.

8
0

Re: Still warming, but less so

"but only about half what's been claimed"

Half using all Class 1\2 stations.

If you use rural stations excluding airports, the number is nearly three times smaller (+0.108 per decade).

Bear in mind that the study period is 1979 to 2008, so in 29 out of 30 years of the study the PDO was in positive (i.e., naturally warming) mode. Considering our findings, there is some small amount of wiggle room for AGW during this period, not not much, really.

7
1
Silver badge

Quoting George Carlin

"The temperature at the airport is 87 degrees. That's stupid, I don't know anyone that lives at the airport! Downtown is much hotter! This just in... Downtown's on fire, man!" - Your hippy-dippy weatherman

1
0

Re: Quoting George Carlin

Yes, the absolute temperatures are cooler in airports. But the temperature TRENDS (sic) in downtown urban and airports are much the same (i.e., each is exaggerated by c. 0.1 C per decade).

1
0

Keeping thermometer next to the incinerator

DOES increase the temperature, who would've guess?

11
1

Re: Keeping thermometer next to the incinerator

Well, to be fair, the question is whether it increases the TREND.

And that was anyone's guess.

3
0

If the findings of the Watts paper are found to be correct, I think this is of great significance. I'm not a big fan of WUWT, but I read this paper, and it makes its case clearly. I've seen his contributions dismissed in the past because he is a "mere weatherman". In this case, as the topic is weather measurement, he can't be dismissed that easily.

10
1
Bronze badge

From what I gather, the authors of the study are suggesting that temp data taken from locations near a ton of concrete/asphalt etc, are skewed upward. Fair enough, but it would be nice if the .pdf had actually included the method of calculation in an appendix. Things like heat transfer equations, constants used (such as assumptions on wind affecting convection rates, thermal capacities etc). I notice these things are all nicely referenced, but as usual us average people need to take a trip to the nearest university library to check those references. I'm curious just not that curious. I realize this is just the "pre-publication" draft, but if these guys went the extra mile to dump everything in their "big-ass climate study" folder out in the open, it would do wonders (in my opinion at least) for the quality/credibility of the work.

2
0

Reference

KS, I haven't had time to read the .pdf paper yet, but the press release does include this reference to Leroy's updated rating method:

Leroy, M., 2010: Siting Classification for Surface Observing Stations on Land, Climate, and Upper-air Observations JMA/WMO Workshop on Quality Management in Surface, Tokyo, Japan 27-30 July 2010 http://www.jma.go.jp/jma/en/Activities/qmws_2010/CountryReport/CS202_Leroy.pdf

2
0

Not accurate

The authors of this study aren't suggesting that; that's the urban heat island effect, which isn't at all controversial at this point -- it can be seen in unadjusted modern weather station data. The study under discussion here contends that Hansen used measurements from within urban heat islands, which are well known to be higher than measurements taken elsewhere, as the baseline for normalizing weather station temperature data in general; the contention here is further that, urban heat islands being a very small percentage of Earth's land area, to use them as a baseline for normalization produces a warming trend that's skewed upward by a factor of two in comparison with the result from a normalization weighted for the enormous difference in land area between urban heat islands and everywhere else.

5
0
Bronze badge

Re: Not accurate

Thanks for the correction, I should have said "..skewing the overall data trend upward.."

3
0

Leroy (2010) methodology is indeed included. It is in one of the links accompanying the paper.

It is also carefully explained in the PowerPoint sheet.

And we are not measuring temperatures. We are measuring temperature TRENDS only. I cannot emphasize this strongly enough.

Anthony cites a number of papers examining heat transfer, etc., in the paper itself.

What we do is rate the stations for heat source/sink proximity are area coverage using Leroy (2010) methodology. The paper is not trying to find out WHY there are differences. The paper is trying to determine IF there are differences and how great those differences are.

3
0

Page:

This topic is closed for new posts.