[Ian Williams]

19 March 2012 Last updated at 16:37
Update for world temperature data
By Mark Kinver Environment reporter, BBC News The Arctic is warming faster than any other region on the planet, data shows Continue reading the main story
Related Stories

• Arctic climate 'tech fixes' urged
• Ice 'tipping point' may not occur
• New warning on Arctic ice melt

Researchers have updated HadCRUT - one of the main global temperate records, which dates back to 1850.

One of the main changes is the inclusion of more data from the Arctic region, which has experienced one of the greatest levels of warming.

The amendments do not change the long-term trend, but the data now lists 2010, rather than 1998, as the warmest year on record.

The update is reported in the published in the Journal of Geophysical Research.

HadCRUT is compiled by the UK Met Office's Hadley Centre and the Climatic Research Unit (Cru) at the University of East Anglia, and is one of three global records used extensively by climatologists.

The other two are produced by US-based researchers at Nasa and the National Oceanic and Atmospheric Administration (Noaa).

Cru's director, Phil Jones, explained why it was necessary to revise the UK record.

"HadCRUT is underpinned by observations and we've previously been clear it may not be fully capturing changes in the Arctic because we have had so little data from the area," he said.

"For the latest version, we have included observations from more than 400 (observation) stations across the Arctic, Russia and Canada."

Prof Jones added: "This has led to better representation of what's going on in the large geographical region."

Despite the revisions, the overall warming signal has not changed. The scientists say it has remained at about 0.75C (1.4F) since 1900.

Kicking the bucket

Another change adopted in the HadCRUT dataset is the way sea surface temperature (SST) is recorded, allowing scientists to revisit and recalibrate past calculations.

With advances in technology in recent years, ships now have electronic sensors that can accurately record SST.

This development has highlighted a systematic anomaly in traditional methods of collating the data in the past.

This included differences in the buckets used to collect sea water for measurement, and the locations where those measurements were recorded.

Improvements in the way SST is collected has now allowed scientists to recalculate data, making amendments to the data collected in previous years.

"An example of this is the rapid change in the kinds of measurements we see in the digital archives around the Second World War," explained Peter Stott, head of climate monitoring and attribution at the Met Office.

"Research has shown readings from buckets were generally cooler so when the database changes from one source to another, you see artificial jumps in the temperature.

"We have quantified these effects and corrected them, providing a clearer view of the evolution of global temperatures."

http://www.bbc.co.uk...onment-17432194

This post has been edited by Ian Williams: 20 March 2012 - 15:12

[Flatlander]

Yeah, they made 1998 colder.

[John Mason]

Comes into line with most other global temp records in terms of trendline. HadCRUT3 was missing scores of high-latitude stations' observations, so it was less representative, and hence a bit out-of-step with the others, which included those data. Need to find time to have a more in-depth read of all this, though.

Cheers - John

[Wiseacre]

In what other science do the measured variables keep changing depending on someone's interpretation? If the figures are suspect or the method of collection the same, they should be eliminated from the calculation. Furthermore, if such 'altered' data is used, the resulting summary or graph should make this very clear.....that it is an estimate and not a fact. Until the science of this subject is respected by the scientists, many will struggle to subscribe to this religion.

[Andy Mayhew]

Hmm, I think the science is very much respected by scientists - its certain other people who don't respect it because they don't understand how science works.

As an anology, if you had data from 10 stations in England and used that to determine temperature change across Britain, and then found data from 200 stations across England, Scotland, Ireland and Wales, would you change your temperature record to reflect the new data, or stick with the old?

[Uskys]

Andy Mayhew, on 23 March 2012 - 21:40, said:

Hmm, I think the science is very much respected by scientists - its certain other people who don't respect it because they don't understand how science works.

As an anology, if you had data from 10 stations in England and used that to determine temperature change across Britain, and then found data from 200 stations across England, Scotland, Ireland and Wales, would you change your temperature record to reflect the new data, or stick with the old?

Andy , Thats rather patronising.

Wiseacre posted a very salient comment. Your analogy is not remotely applicable. Data has been altered.... that is, temperature recorded by calibrated instrumentation has been modified, altered, interpolated, distorted etc ... for what purpose? thats the question.

[Wiseacre]

Tsk tsk, Andy. You do need to read the post before commenting. Your analogy would be correct if I was talking about the added high latitude stations, but I was refering to the continual 'adjustments' made to HadCRUT (for example to counter assumed UHI).

This post has been edited by Wiseacre: 23 March 2012 - 22:26

[ldavidcooke]

Hey All,

What everyone is touching on is a basic issue with what the Global Average Temperature is supposed to represent. Is it on the one hand supposed to represent the air temperature that "everyman" is supposed to encounter as they open the front door to get the paper/mail/let out the dog/cat without spilling their tea or dropping their scone/banger? Or on the other hand is it a case in that science wants the GAT to represent the intermediate average air temperature at roughly 1020mb alitude or 2 meters above Sea Level?

The problem is, in the first case the dominat fauna has de-natured the environment; which means if you were to remove them from the species on the surface of the planet, the GAT should drop 6 deg. C within 100 years. In the second case, the earth does not have a uniform albedo/composition; hence, there can be no uniform 1020mb GAT.

That science is attempting to derive a uniform surface and distribution of thermal content leads to a modification of the actual ground measure in an attempt to simplify the thermal dynamic equations describing the thermal gains and losses of the planet.

The result, two different record/data sets are met with two different applications. In short, choose your preference and do not mix references and we are all good.

[Andy Mayhew]

My apologies, I think I've been spending too long looking at certain other forums where disingenuous attacks on science by 'Luddites' are commonplace and rather jumped in there without thinking ... I agree that on the face if it it seems strange to adjust existing data, but there are a number of papers on the subject explaining how and why this is done. I'll see if I can dig some out later.

[Andy Mayhew]

Peter Stott's explanation of the changes for HadCRUT4 can be read here:

http://www.metoffice...pt_HadCRUT4.pdf

I've not seen a paper on the latest dataset, but one referring to the previous one (HadCRUT3) is here:

http://www.cru.uea.a...T3_accepted.pdf

There's also a useful little explanation from the the NZ Met Office on why they adjust their data here:

http://www.niwa.co.n...-to-be-adjusted


Basically, it's all about trying to get the most accurate like-for-like data record. Stations change, equipment changes. And I'm sure that sometimes, after adjustments have been made, someone goes back over the paper produced showing how and why they were changed, and finds something new that was missed or not calculated correctly. Science never stands still. It's always changing, we're always striving to improve the knowledge and data we have. We never say 'if it was right in 1990 then it must be right today'.

(as an aside, I see they still can't decide how high Mount Everest is - and that should be a lot easier to claculate than 150 years worth of global temperatures)

If I were to produce an accurate temperature trend record for my back garden I would have to adjust most of my data because over the years I have used different sensors in different positions and I'm quite certain that the equipment I use today would have recorded slightly different temperatures 8 years ago compared to the (less accurate) equipment I had back then.

I am quite certain that, if after adjustments have been made, the warming trend (for example) was less pronouced than before, it would still be published - and, perhaps cynically, that some people would be much less inclined to object to such adjustments.

[ldavidcooke]

Hey Andy,

However, the current adjustments are not the 0.5 deg difference in accuracy when measuring with a Mercury themometer versus a Bi-Metal Spring meter or a Digital thermal (pyro) sensor. Each has their own error bars due to the instrument employeed.

That is not the issue here, the issue is when you attempt to move from a known value to an, for all attempts and purposes, imaginary value. Trying to sell the "adjustment" to the average "Joe" is never going to fly. To go further to suggest to use that imaginary value as an absolute proof of AGW is going to be little different then trying to pass the proverbial camel through the needles eye, it just won't sell. Especially if you attach taxes based on the imaginary numbers...

The best sell is to say todays temperature in downtown Sheffield, on Church St. is a average of 15.5C, where as 45ya it would average 11C. The differences could be attributed to many things; however, the average today is indeed higher then in the past, period.