The scandal of fiddled global warming data

[cybercoma]

Oh man. *facepalm*

[On Guard for Thee]

Oh man. *facepalm*

Double facepalm!

[TimG]

Adjusting is part of the science. The deniers always celebrate when the temps are adjusted their ways, so...

I have two issues with adjustments:

1) Data is only ever adjusted in ways that enhance the alarmist cause. It is never adjusted in ways to diminish the cause. This is because the people tasked with managing this data only care about errors which make it seem less severe so they hunt for them and exaggerate their significance. They simply don't care about errors which make it seem worse than it is. In fact, it is unlikely that any adjustments that reduced alarm would ever be accepted by scientists working in the field because almost none of them want to "hurt the cause" by publishing such adjustments.

2) The comes a point when the data has been adjusted so much that it is no longer data and becomes an artificial construct. Whether this construct has any credibility depends on whether you trust the people doing the adjustments. I don't so most data that has been subject to multiple adjustments over time is junk in IMO. It is worse than having no data at all because the false illusion of knowledge is worse than the truth of no data.

PS. if you disagree with 1) then I challenge you: find one climate dataset where the net adjustments (note the word "net") diminished the cause for alarm. I say such an example does not exist but I can't prove a negative.

Edited June 25, 2014 by TimG

[Mighty AC]

You just admitted it again. No cite required.

Ouch! Quit while you're behind.

[Michael Hardner]

I have two issues with adjustments:

1) Data is only ever adjusted in ways that enhance the alarmist cause. It is never adjusted in ways to diminish the cause.

Never ? Didn't the coefficient just get adjusted downwards by the MET office in the UK ? And didn't that reverberate in the anti-climate press there ?

2) The comes a point when the data has been adjusted so much that it is no longer data and becomes an artificial construct.

At a certain point, yes, the coefficient would be reduced to the point where it blends in with 'noise'. Are we close to that point today ?

http://en.wikipedia.org/wiki/Climate_sensitivity

I don't think the climate sensitivity ranges have been adjusted yet, so the answer is it's possible but not currently happening.

----

And - as someone who raises (to my mind) reasonable questions about climate science, I believe it's your duty to point out the ridiculous viewpoint that Bryan raises, ie. any adjustments equate to lying.

Of course, if you consider him to be on your 'team' then you'll shut up and let the quality of discussion suffer.

[TimG]

Never ? Didn't the coefficient just get adjusted downwards by the MET office in the UK ? And didn't that reverberate in the anti-climate press there ?

I don't think you really understand what we are talking about. The MET acknowledged that, despite their attempts to adjust the data to show more warming, they were not able to adjust it enough to eliminate the "pause". They also assured us the the "pause" is temporary and warming continues. This is NOT an example of adjustments to data making it seem less severe - this is an example of data being so against the meme that they were forced grudgingly acknowledged it. The "pause" was evident to skeptics back in 2008 but no one wanted to listen.

At a certain point, yes, the coefficient would be reduced to the point where it blends in with 'noise'. Are we close to that point today ?

We have surpassed it today. Most of the warming in the surface data comes from 'adjustments' that have a superficial justification but really on the judgement of the data custodians. For example, the TOBS (time of observation) adjustments are justified from a scientific perspective but it is highly suspect in practice because these adjustments greatly increase the observed warming in the complete dataset when random chance suggests that the TOBS bias should be randomly distributed across the entire dataset even as it biases individual stations.

I don't think the climate sensitivity ranges have been adjusted yet, so the answer is it's possible but not currently happening.

Sensitivity is not data. It is calculated value that depends on many assumptions.

And - as someone who raises (to my mind) reasonable questions about climate science, I believe it's your duty to point out the ridiculous viewpoint that Bryan raises, ie. any adjustments equate to lying.

The guy who made the original allegations has been taken to task by skeptics for that claim. Watts and few other prominent skeptical bloggers make it clear that they do not think the data custodians are actually fabricating data or knowingly lying. That does not mean they believe the adjustments are justified:

http://www.climate-skeptic.com/2014/06/my-thoughts-on-steven-goddard-and-his-fabricated-temperature-data-claim.html

Edited June 25, 2014 by TimG

[waldo]

For example, the TOBS (time of observation) adjustments are justified from a scientific perspective but it is highly suspect in practice because these adjustments greatly increase the observed warming in the complete dataset when random chance suggests that the TOBS bias should be randomly distributed across the entire dataset even as it biases individual stations.

I have little time today to deal with your complete nonsense... perhaps I'll revisit later this evening. For now, I'll speak to your TOBS sleight-of-hand play. Somehow you conveniently miss the salient background to that adjustment which, of course, has to do with the time measurements are made relative to the 'midnight reference hour'. Within the U.S., for example, prior to 1940, per a ruling guideline, measurements were taken near sunset... however, over time (for operational concerns) there has been a gradual shift in when measurements are taken - a wholesale change to taking measurements in the morning. These "morning time of observation measurements"... given lower temperatures in the morning... have acted to introduce a bias, a broad-scale reduction in mean temperature. The TOBS adjustment properly eliminates this reduced mean temperature bias... the adjustment methodology is sound and has been subject to the scrutiny of several published papers.

[cybercoma]

At a certain point, yes, the coefficient would be reduced to the point where it blends in with 'noise'. Are we close to that point today ?

That's not really how it works. The size of the coefficient has little to do with its statistical significance.

[Michael Hardner]

That's not really how it works. The size of the coefficient has little to do with its statistical significance.

Right... because the function would still change over time just more slowly. Ok.

[Michael Hardner]

The guy who made the original allegations has been taken to task by skeptics for that claim. Watts and few other prominent skeptical bloggers make it clear that they do not think the data custodians are actually fabricating data or knowingly lying. That does not mean they believe the adjustments are justified:

Thank you.

[cybercoma]

Right... because the function would still change over time just more slowly. Ok.

The effect is just larger or smaller, but can still be statistically significant. If it lacks statistical significance (blends in with the noise as you called it) then the scientists would state that there is no statistically significant effect, despite what the coefficient actually is (it could be large or small).

Edited June 25, 2014 by cybercoma

[Topaz]

Question...do any of you think that HAARP has anything to do with the weather or is that another tinfoil, conspiracy subject?

[Bonam]

Question...do any of you think that HAARP has anything to do with the weather or is that another tinfoil, conspiracy subject?

HAARP is a scientific program to understand ionospheric phenomena and their uses/relationships to communications as well as other basic scientific research, which you can read about in numerous papers. It's also in the process of being shut down this year.