Global Air Temps

It is difficult to the point of being almost impossible to measure small but significant global temperature changes by directly taking temp readings of any single thing.

Much better evidence exists by observing the indirect effects; glacier melt and plant growth pattern changes and animal migrations.

 

I don’t think you have Salby’s result right. He showed that CO2 growth rate is proportional to temperature.
So if temperature becomes constant, then CO2 should grow at a steady rate. Looks like it does;
see his talk at the British Parliament (via Bishop Hill).

 

It was the IPCC that gave a range of temperatures from the last 15 years period may have been cooling
http://www.ipcc.ch/pdf/assessment-report/ar5/syr/SYR_AR5_SPM.pdf

In there confidence interval for the 15 year period temperature models disagree, and there could have been a little cooling, or a little warming, or maybe nothing. Statistically I'd have to say no significant warming or cooling. If you buy the explanation, ENSO, which I do, then you have to figure out how ocean oscilations will affect temperature in the future to predict how much warming we will get.

Now you can take that no statistical change and draw different lines and say we are cooling. IMHO that would be bad statistics, but there are lots of bad statistics in public policy. Now 1998 is an unfortuanate date to pick, as an end point, and you could blame the skeptics of cheery picking it, but really the IPCC picked it in the third assessment as part of the statistical graphting that made the temperature record seem so spiky.

 

Yep it was an outlier when the IPCC published 3. Instead of handling it like an outlier the summer acted as if it was part of a trend, which statiistically you can't tell in climate until at least 15 years in the future. Oh damn, we are here and it was indeed an outlier.

There are easy statistical methods for handling outliers, but IPCC does not seem to want to use them. Many papers though individually include them. The most common is to remove them from the averages, bur discuss them.

Absolutely, a wonder full set of questions starting with

1) How do ghg and temperature affect ENSO and other oscillations
2) How do ENSO and other oscillations change temperature.

 

the data and model define outliers. Here I was refering to IPCC models. 1998 clearly is an outlier to the models.
1.3.5.17. Detection of Outliers

Here the outliers probably define something interesting that is missing from the models and not bad data. The 5th assessment theorizes this is ENSO. I am not sure, but clearly all ocean ocilations should be considered as possible reasons as well as changes in ocean temperatures.

My problem is in the modeling. If you model without considering outliers, you likely will badly predict future tempeatures. We see the models presented in IPCC 3 badly predicted temperatures, so the results fit the theory. If we continue to model in the same way as in IPCC 3, there will continue to be systemic errors in climate prediction.

It should have been called out as an outlier and modeled differently in the predictive models. 3 standard deviations is not a majic number in statistics, but certainly if something were 3 sigma off of a model something is very off from that model.

I would throw it out or at a minimum smooth the deviation away, if you are mainly handling S log (ghgY/ghg X) where S is the sensitivity ghgY is concentration in Year Y, and ghX is concentration in reference year X. If the model is more complex than perhaps it should be included.

 

FYI, others have come to the same conclusion:

THE HOCKEY SCHTICK: New blockbuster paper finds man-made CO2 is not the driver of global warming

Swedish scientist replicates Dr. Murry Salby’s work, finding man-made CO2 does not drive climate change | Climate Depot

 

Skimmed the article, I'll ad quotes from publicly available.
Alfred Wegener Institute Detail

First I wouldn't blame the modelers here. The consensus view has been that the majority (over half) of temperature variation is due to ghg (carbon dioxide, methane, sulfur dioxide, nitrous oxides, particulates bothl ight and dark). The majority of models will go there as it follows what the modelers believe the science to be.

Certainly data gets vary sparse older than 1400 years ago, so models rarely go back further than this for testing, and most tests are only of the temperature record period.

So the problem with the models is A) assumption of majority ghg (although this may still be correct some should be made with 0 or negative feeedback on ghg, B) Bad data to model on, C) no good theories to model ocean ocsilations to ocean heating and sst.

Models seem to assume that sensitivity to ghg is constant, but this is only likely true for bounded time periods. Sensitivity may be different when ghg leads temperature than when it lags it, and it is likely to be different before and after tipping points. Therefore I would not use these models to go back past a tipping point (was there one when the mwp switched to lia?, if so only train with the last 500 years), if not whatever caused the end of the mwp needs to be modeled.

 

It's not denial to question what you are told, but the basis for questioning this assertion are many. Assertions based on other assertions are suspect to begin with, but additionally, past statistical weather information has been manipulated and even destroyed, thereby making it all the more imperative to be asking questions.

 

Although we have emails from phil jones that he intended to destroy some backing information at the CRU, we have 3 other sources of temperature reconstructions from instrument data NASA GISS, NOAA NCDC, and Berekely B.E.S.T.. All of these are in broad agreement, all are us, 2 from the US government, 1 from a state university with public and private grants, even koch brothers money used to do the reconstructions. Do you have any evidence that any of the US based reconstructions manipulated or destroyed data?

It is the CRU data from 1998 which is the highest. If you use that data, a decline of temperature of 0.05 per decaded for the last 15 years is within the margin of measurement error. The other three have lower reconstruction for 1998 so have a slightly higher temperature now than in 1998. Most skepticism is on the predictive models which have done a poor job lately.

 

That last line is a poor generalization of climate skepticism, but more importantly, the quality of an assertion is not diminished by the intensity of the skepticism applied to it, but rather the degree to which it has held up to it.

Anyone who considers themselves a scientist (myself included) should be ashamed by the notion that those who challenge a scientific assertion, are themselves out of order in some way, and deserving of a pejorative label like "denier."

 

Greg,
no problem with you being skeptical, my problem is skepticism because of you haven't looked into the science and are just repeating hearsay. So please look at NASA GISS, NOAA NCDC, and Berekely B.E.S.T. and tell us if you still distrust the temperature record and why. Certainly if your conspiracy theory is they are faking it, why would we have had little to no significant warming since 1998?

I think I have been called denier on this board many times, and it terms out on most of them the IPCC has reversed itself or at least is more open to a skeptical interpretation. The temperature record though is pretty golden in the last 50 years, and we have a good idea of how big potential errors are before then from 1880. B.E.S.T. even researched most of Watt's objections to the records at the other places (urban heat island, methods of calculating sparser past data, etc). The data sets and methods are public there.

My big ovjection is to the religionist. Those that claim anything in the IPCC's bible is true forever, or the oposite - people like Inhofe. Inhofe says climate change is a hoax because his bible says man can't do anything to change the climate. That is assinine.

 

Religionist, politicist, activist; none of these words square with "scientist." When science is contaminated by these attitudes, it's not science at all, its quite the opposite.

 

Our good friend Austin is suggesting that finding and posting original papers with new data is well respected here. Postings from an advocacy site are less useful because they often omit counter data and the poster of this 'interpreted' seldom carries any critique back.

Many of us here are interested in facts and data. Bring some from the original papers and Austin, I, and others will treat the post and poster with respect. But reposting advocacy summaries really doesn't work . . . or trying to make it political . . . that is ineffective.

Bob Wilson