Ronaldo, which would your grandfather have thought made less sense:

- using models to predict the future programmed by people who could not explain the past?

- using models to predict the future whose only 'testing' consisted of a check they could reproduce the data used to parameterise them?

"The best current description of Holocene temperatures is Marcott et al 2013.
Sep 9, 2014 at 1:17 PM | Unregistered CommenterEntropic man."

Gimme a break, EM.
Marcott produced another 'Hockey Stick' where the 'blade' magically appeared in the short time between when he presented his work to his Ph.D. committee and when he submitted it to Nature. Under questioning from Steve MacIntyre he admitted the blade was "not robust".

As for the rest of the "stick", who believes a global temperature recontruction that rests largely on a mixed-bag of paleo-proxies taken from coastal regions around the continents at a time of rapid sea level change?

If Marcott et al 2013 is the best representation of the Holocene, then you have just damned a lot of more careful scientists.

It didn't.

The Mediaeval Warm Period appears as short term variation on the long term decline from the Holocene Optimum.

When you look at climate data several types of variation should be considered.

One is internal variation, the randomness that comes with weather, volcanoes etc and adds noise to the record without predictability. Though its effect cannot be forecast its effect on confidence limits can be predicted and allowed for.

Second is external variation due to solar cycles, PDO, ENSO orbital changes etc. Which create predictable oscillations or long term trends which emerge from the data once the time series is long enough to separate the trend from the noise.

Note that both of the above are natural variation, driven ultimately by the physics. Under natural conditions CO2 is part of this, amplifying temperature changes driven by other factors.

Anthropogenic carbon release is similar to your computer. Without a human civilisation the 400ppm CO2 would not exist, nor would the temperature rise it is causing. Without humanity, left to natural variation CO2 would be about 280ppm temperatures would now be about 1C cooler and declining gradually.

"..adds noise.."
Sep 9, 2014 at 3:39 PM | Unregistered CommenterEntropic man

Amen.

"Without a human civilisation the 400ppm CO2 would not exist, nor would the temperature rise it is causing". -
Entropic man. at 3.39 pm.

"Everyone** agrees that we can't predict the long-term response of the climate to ongoing CO2 rise with great accuracy. It could be large, it could be small. We don't know".
Professor Richard Betts BH Blog 24 August

Compare the certainty of EM with the acceptance of reality by Prof. B

Come on chaps, sort yourselves out.

Martin A

Martin A

Without confidence limits or statistics how do you judge the quality of your data?

When you engineering has to take into account every possible variable, important or minor, why are you not paralysed by your own uncertainty? How do you design a circuit when your philosophy requires you to use the full Schrödinger wave equation, when Ohm's law would suffice for the rest of us?

EM
Sep 9, 2014 at 7:44 PM

In reply to my question about the appearance of the MWP
you replied - "it didn't"

Have a look at the wealth of citable evidence for its global existence provided in the following ref. and think again about eg. the arrival and departure of the Vikings from Greenland.

http://www.co2science.org/subject/g/globalmwp.php

While you are thinking, have a look at the Roman Warm period.

Ronaldo

We are talking at cross purposes, I think.

You see the RWP and the MWP as separate events. I see the as part of the larger Holocene temperature pattern.
From 8000 to 4000BP global temperatures held around 14.6C and then started a slow decline of about 0.6C to the late 1800s. The warmth around the RWP and MWP shows as an upward blip in the frst millennium AD and then as a downward correction from 1100AD, the LIA.

From 1880 we have the observed temperature record which shows a very rapid temperature rise, way outside previous Holocene norms.

I fail to see the distinct boundaries in the ensemble data which you see in individual proxies. I have no problem with the argument that it was warmer 1000 years ago than in 1880. I do have a problem with the sceptics who claim that the RWP/MWP were warmer than today.

Rinaldo

Please give detonate references. You link is an example of a Gish Gallop .

Michael hart

Curiously, all the proxies with coverage over the last 1000 years produce hockey sticks. It is not just the tree rings.

"Curiously, all the proxies with coverage over the last 1000 years produce hockey sticks. It is not just the tree rings."

a bizarre example of hyperbole considering how the majority of tree rings do not obey this rule, nor do the Finnish lake sediments. maybe EM can offer some proof. Or is this one of his feints - eg there are 3 proxy series with 1000 year coverage and they all show an uptick in the 20th c. Sadly, all the other proxies do not follow the same course. Watch the pea. Or is it only 1000 year proxy series that should count in this analysis, in which case, he should get in touch with Mann and Marcott.

EM
Please explain your last post addressed to Rinaldo, I genuinely do not understand it.

I suspect that your previous reference to a "cross purposes" discussion results from you relying on Marcott's use of a 400 year averaging filter whilst I am looking at the fine structure of the ice core data.

The main reason why the topic gets debated at all in this context is that while there appears to be nothing particularly exceptional about current warmth in the history of human civilisations, the IPCC alarmism requires it to be "unprecedented". Smoothing, whether done mathematically or by use of proxies with poor resolution, hides the extremes from the past. By using unsmoothed present day 'weather noise' and 'smoothed climate' data from the past, it is child's play to produce hockeysticks.

What are the odds that EM will now probably attempt (again) to reverse the normal order of scientific hypothesis testing and insist that sceptics have to prove that present day warmth is not as warm as the Medieval, Roman, and Minoan warm periods?

EM:

"Without a human civilisation the 400ppm CO2 would not exist, nor would the temperature rise it is causing. Without humanity, left to natural variation CO2 would be about 280ppm temperatures would now be about 1C cooler and declining gradually."

You can not possibly know that!

Have you taken into account:

Land use changes caused by humans?
Coastal reclamation caused by humans?
General industrial pollution caused by humans?
Draining of low lying land?
Large scale farming?

Each will have an effect increasing CO2 or decreasing, sometimes measurable, sometimes not.

To say that without humans the earth will be a different temperature currently must be: a guess.

Michael hart

Why would I want you disprove that it is as warm now as in the past? We can both read temperature data and it shows that current conditions are warmer than any of the periods you mentioned.

From Marcott et al 2013 Fig 1b and the current GISS temperature record I got the following figures.

For the Minoan Warm period circa 700BC the temperature anomaly was +0.07C +/- 0.2C

For the Roman Warm Period circa 325 AD the temperature anomaly was -0.05C +/- 0.2C

For the Mediaeval Warm Period circa 1100AD the temperature anomaly was -0.19C +/- 0.2C

The temperature anomaly for 2013 was 0.62C +/- 0.09C.

Ronaldo

When an ice core forms, there is considerable vertical movement of water and other solutes within the firn before it is completely frozen. This smears the years together enough that the fine detail is mostly noise. Most other proxies have the same problem. Your attempt to use fine detail to make climate judgements is trying to turn noise into signal and is unsound.

Steve Richards

I projected the graph mentioned above, assuming that the long term trend to 1880 continued, without the post-1880 anthropogenic warming. Consider it a hypothesis.

We can both read temperature data and it shows that current conditions are warmer than any of the periods you mentioned. Sep 12, 2014 at 8:36 PM | Unregistered CommenterEntropic man

Not by the same yard-stick they aren't. They calibrated their proxies to one of Mann’s constructions, not GISS. (So perhaps no surprises there when a hockey stick appears.)

We can both read temperatures all right, but there weren't any to read back then. As you well know, there are no direct temperature data for those periods that can be properly compared to the present. That is why people are reduced to using proxies, not thermometers; picking ones they like and discarding others that cause problems; maybe re dating some core tops:
http://climateaudit.org/2013/03/13/marcott-mystery-1/

As I was trying to explain, Marcott's proxies also cannot give you the detailed high frequency data. But that doesn't mean large short-term temperature swings didn't happen in the past. Figure 1a and 1b show the mean (plus only one sigma uncertainty. I'm not sure that I ever read if anybody understood how they calculated their uncertainties), not the extremes that make up the average. RSS satellite temps shows variation around 2011-2013 alone to be more than 0.5 degrees What does that prove? Not much, except that I can cherry-pick just as well as you or Marcott can.

I already said above that the authors conceded the 20th century data was "not robust". In more detail they said:

"20th century portion of our paleotemperature stack is not statistically robust, cannot be considered representative of global temperature changes, and therefore is not the basis of any of our conclusions."

“The resolution of our reconstruction limits our ability to determine if the recent warming of the past few decades has reached maximum interglacial values; however, from model-based projections of 21st-century climate we infer that the mean global temperature will rise well above the warmest levels of the current interglaciation by 2100.”

Maybe you should write and tell them you've found a better way.

----------------------------------------------

Why would I want you disprove that it is as warm now as in the past?

I didn't ask about your motives. I'll leave that to you and your conscience.

Michael Hart

Horses for courses. Marcott et al is a reconstruction of a long term trwnd , not a tool for studying short term ie decadal variation.

Marcott et al used a method of analysis with a temporal resolution of 120 years. That would not have shown, for example, the decrease in rate of change from 1940 to 1970.

It is good enough to show the Minoan, Roman or Mediaeval warm periods or the LIA if they existed as separate entities.

The confidence limits are standard statistics, as would be used for any numerical data.

The Marcott et al 20th century data has caused considerable sceptic angst
mostly for no reason.

1) Fewer proxies are available as you get closer to the present. This makes the 20th century data less reliable. The choice was to run the analysis as far as it worked and point out the decline in reliability, or choose some arbitrary cutoff date. Which would you have done and, if you chose the cutoff, what date would you have chosen and how would you have justified it?

2) There is another source of temperature data. We have 134 years of measured temperatures. It was interesting that the proxy data and the record showed similar behaviour but not unexpected and no reason for hysteria.

I keep encountering sceptics who wave past warm temperatures like a sword. Who stated that silly meme, and why do they reject Marcott et al, which confirms that it was warmer in 8000BC than in 1800AD?

Martin A

Lots of hypothesis testing going on and it supports AGW. In the context of a fringe climate sceptic website you can be as free as you like, but out in the scientific world the AGW paradigm is the one that fits the evidence best. If you want it replaced by lukewarmism, skydragon slaying or leprechauns ( name your preferred alternative) you have to demonstrate that the available evidence is better explained by your paradigm than theirs.

I have heard so much false criticism of climate science by sceptics that it no longer passes my bullshit filters.Nit picking that this was done wrong or that could have been better is no help. Despite all the sceptic flak most of the work is competently done and relates to the real world. To change a paradigm you dont tell the practitioners they are incompetent when they know their competence. You change the way in which the existing evidence is viewed. You do that by showing that your way works better.

Anything else is politics and propaganda.

Marcott et al used a method of analysis with a temporal resolution of 120 years. That would not have shown, for example, the decrease in rate of change from 1940 to 1970.Sep 14, 2014 at 3:33 PM | Unregistered CommenterEntropic man

Well, congratulations. Dragging that kind of honesty out you seemed like squeezing blood out of a stone.

The confidence limits are standard statistics, as would be used for any numerical data.

There is no such thing. It must be clear how you calculated something. Applying correct mathematics to the wrong data would be unacceptable.

The Marcott et al 20th century data has caused considerable sceptic angst
mostly for no reason.

Angst? Hardly. The author effectively retracted it, as I detailed above, and you studiously ignored for the n'th time.

Dismay? Yes. Dismay that supposedly competent, impartial and independent sources would allow such a wilfully deceptive article to be published. It was another hockey-stick in the Michael Mann mold.

1) Fewer proxies are available as you get closer to the present. This makes the 20th century data less reliable. The choice was to run the analysis as far as it worked and point out the decline in reliability, or choose some arbitrary cutoff date. Which would you have done and, if you chose the cutoff, what date would you have chosen and how would you have justified it?

No. They just diverge significantly and or produce the wrong results. That is why Michael Mann chose to ignore tree rings from the mid 20th century in the original hockey-stick. Marcott selected the proxies he wanted, redated some, ignored others, got the result he wanted, and then admitted it was "not robust" when called on such BS.

2) There is another source of temperature data. We have 134 years of measured temperatures. It was interesting that the proxy data and the record showed similar behaviour but not unexpected and no reason for hysteria.

Two answers. No, and No. Three actually. The third one is also No (unless you count the global-warmists hysteria, but that is ever present. And we've got a lot more than 134 years of measured temperatures, depending on location.

I'll say it again: The modern temperature record will show greater temperature swings than the proxy record because the proxy record is incapable of recording such temperature swings. You've as good as admitted it above. Get a grip.

Michael hart

Oh dear, I think I've misunderstood. Do you regard the Mediaeval warm period, the Roman warm period and the Minoan warm periods as transient events lasting less than the 120 year detection threshold of Marcott et al? .

If so, what do think temperatures were like between these transients? On what evidence? What lesson do you infer about 20th and 21st century temperatures?