Buy

Books
Click images for more details

Support

 

Twitter
Recent comments
Recent posts
Currently discussing
Links

A few sites I've stumbled across recently....

Powered by Squarespace

Discussion > A single repository of scientific scepticism

It is the calculation that has the wrong units, which you'd understand if you read the article.

Feb 26, 2016 at 12:47 PM | Unregistered CommenterRaff

Essex and co said "R4 would appear in connection with black body radiation" which might be construed as confusing when they then applied the measure to Celsius temperatures as NIV agreed.

But apart from that I can't see anything wrong. I did a quick plot myself of the curves and visually could not spot anything different from the Essex et al curves. Rabett is just dragging a red herring and I think you followed the trail he laid. They were not doing radiation calculations and it is misleading of Rabbet to imply they were.

Essex et all chose a temperature scale and showed the result of their choice. They themselves said that different scales could have been used.

Feb 26, 2016 at 2:36 PM | Registered CommenterMartin A

I also tried pointing that out, in vain, Martin A.

Raff apparently didn't even read my reply to aTTP who made a passing comment about not seeing any T^4 numbers in the article, when in fact it is one of the main explicit points made by Essex McKitrick Andresen:

"It is not in global thermodynamic equilibrium — neither within itself nor with its surroundings.
It is not even approximately so for the climatological questions asked of the temperature field.
Even when viewed from space at such a distance that the Earth appears as a point source, the radiation from it deviates from a black body distribution and so has no one temperature [6].

If Raff regurgitated views he does not understand from else where only once, then we would have cause to be grateful.

Feb 26, 2016 at 3:18 PM | Unregistered Commentermichael hart

Martin, I imagine asking you to read the Rabett is rather similar to people asking me to read the HSI, so I wont bother again. I've forgotten the reason the E&M paper was originally invoked and I've lost interest too. It seems clear to me that only the arithmetic mean has a physical meaning, but if you and michael want to believe that averaging T^4 in Celsius or the many other averages have any worth beyond allowing E&M to write a fatuous paper, I'm happy for you. Go for it!

Feb 26, 2016 at 3:45 PM | Unregistered CommenterRaff

Raff, I did take a quick look through what Rabett had written on the page you pointed to before I posted the above. It is still here on my screen.

Rabbet said

They then calculate the various averages for the coffee and the ice water warming separately up to a room temperature of 20 C. They claim to plot the predictions of the arithmetic average, the harmonic, the RMS and what they call radiation (proportional to T^4 following the Stefan-Boltzmann law). The only one they get right is the arithmetic average (see below). The figure to the left shows what they calculated. I have added the lines showing the temperatures of the coffee and the ice water

I don't know what he is on about. As I said, my curves seem to duplicate those in the Essex et al paper.

If you think the arithmetic average has a physical meaning - you are free to think what you like. I know that previously you pointed out that if you mix equal volumes of cold water and hot water the mix has a temperature that is the arithmetic mean of the temperatures of the two original volumes - the result of the linear relation between temperature and heat in liquids, but that does not mean that an average of temperatures means anything physically in itself. Adding temperatures is like adding exchange rates - the sum of two or more does not have a physical meaning.

Feb 26, 2016 at 6:33 PM | Registered CommenterMartin A

Well I've heard it often enough, but if you say so...

So how would you calculate the "spatial mean of the global temperature field" that you once said you'd be happy with (in place of the term "true global mean temperature"). Would you use arithmetic means or geometric means or mean radiation in Celsius as Essex does?

Feb 26, 2016 at 9:14 PM | Unregistered CommenterRaff

Raff - whatever I may have said once, I would not calculate a mean temperature on the basis that I don't think it tells you anything meaningful. It might be an interesting curiosity but I can't see it as more than that.

I think that climate science latched on to it as a measure of global warming. But now climate science maintains that global warming is there and continuing despite the lack discernable increases in estimates of global average temperatures. So it's far from clear what purpose it now serves.

Generally, if you set out to measure something, it's something you understand physically, you know why you want to measure it, what you are going to do with the measurements and what precision your measurements must have for the result to be useful. Once you have made the measurement, you should know what is the accuracy of the measurement you have just made. I can't see that any of these things apply to global average temperature.

If you did know what you wanted your 'global average temperature' to do, why you wanted it, then you could have a rational basis for deciding what statistic you should compute, how you should weight and combine the measurements you have, what nonlinear processing if any you should apply. For all I know you might decide that some kind of median statistic would be better than an average.

But I think that, if you were setting about it rationally, you would probably decide that one single figure could not meet your requirements and you might then decide to abandon the idea of characterising a very complicated system by the use of a single number.

Feb 27, 2016 at 9:37 AM | Registered CommenterMartin A

So instead of referring to the global mean temperature you'd rather use the rather less catchy "spatial mean of the global temperature field", but you don't know or wont say whether the 'mean' involved is arithmetic, geometric, harmonic, 'radiational' or any other mean proposed by E&M.

Feb 27, 2016 at 3:07 PM | Unregistered CommenterRaff

TheBigYinJames, Martin A

This might be one for the repository.

Roger Pielke Sr wrote a guest post at Climate Etc suggesting that OHC should replace surface temperature as the metric for climate change.

Feb 27, 2016 at 4:56 PM | Unregistered CommenterEntropic man

Great! Now, could you explain how you arrived at this “ocean heat content” figure? That should be interesting, seeing the near non-existence of data to base it on; just a few thousand readings of several million cubic kilometres of ocean, and almost all of them from the more easily accessible portions of it – in other words, there are vast amounts of ocean for which we have absolutely NO data, whatsoever. Methinks that there are an awful lot of suppositions being made, here.

Mind you, as it is the surface of the oceans that drives the weather systems, it does make one wonder why we need to make a wild guess as to what the temperatures … sorry: “heat content” … of the rest of the ocean is. Also, bearing in mind what Martin A has been arguing for quite a while, now, why do we need to know the global average surface temperature, or any global average, for that matter? What will that tell us, when it is the local temperature that influences the weather – e.g. tropical revolving storms require a surface temperature of 25°C; it matters not what the temperature just a few dozen mile away might be, as long as the surface temperature is at least 25°C where other conditions exist, a TRS will develop.

Finally, as I have pointed out before, I suspect one reason for this desire for “ocean heat content” is that more impressive figures can be used – 9.23 x 10^26 being one figure used in an example. But, how does that relate to a temperature? The OHC might be increasing by this scary-looking figure, but how would this be, if expressed as a temperature? How many zeroes would be between the number and the decimal point, and on which side? 0.1°C (or 0.1K, if you prefer) does not look as scary, does it?

Feb 27, 2016 at 9:59 PM | Registered CommenterRadical Rodent

I also note that Mr Pielke snr’s post is not treated too kindly in the comments, either. Interesting.

Feb 27, 2016 at 10:05 PM | Registered CommenterRadical Rodent

So a Thread Titled 'A single repositary of scientific scepticism' has been twisted and manipulated by raff and EM, to get sceptics to work out means of improving their failed climate science.

You don't need to be a sceptic to spot desperation. This is very encouraging for all who come here, suspecting all is not right in climate science.

Feb 27, 2016 at 10:20 PM | Unregistered Commentergolf charlie

EM - I see what you are saying (I think). That the effect of additional atmospheric greenhouse gases is to cause the ocean to absorb additional heat and so the best way to assess the impact of additional GHG's is by assessing OHC. Is that roughly it?

The only problem when you say "OHC is the metric for climate change" is the meaning of "climate change" now becomes "ocean temperature changes that can only be measured by instruments of extreme precision such as ±0.001°C" rather than "changes in the observable weather and climate".

[Having now looked at the posting you point to, I notice it is actually titled "An alternative metric to assess global warming" (my emphasis). To me, that makes more sense than to say OHC is a measure of climate change.

Do you get the point I'm trying to make? I think it's not just me being pedantic. OHC does indeed measure heat accumulation by the Earth. It it does *not* provide any direct measure of the change in the climate (taken as things above sea level that we can actually notice like rainfall, air temperature, how long the frosts last, and so on). So ok as a measure of global warming but NOT ok as a measure of climate change.]

Feb 27, 2016 at 10:21 PM | Unregistered CommenterMartin A

Roger Pielke Sr wrote a guest post at Climate Etc suggesting that OHC should replace surface temperature as the metric for climate change.

Feb 27, 2016 at 4:56 PM | Unregistered CommenterEntropic man

Indeed, EM. I am glad you are rediscovering some science, even if the thought is conveniently late coming to you. It was, and remains, a metric that is probably better than others, albeit less than perfect.

The reasons being that
a) water surfaces can trap a lot lot more short-term heat in the near surface layers than land because of convection transferring heat to the several to many tens, if not hundreds, of metres of surface ocean.
b) the bulk of the water surfaces at these temperatures cover ~70% of the planet
c) the heat capacity of water changes relatively slowly and linearly over the bulk of the regions where the imputed extra heat was expected to accumulate.

Then we have:

d) The alarmist global-warming activists chose to ignore this, and R. Pielke Sr., at their leisurely convenience, while atmospheric temperatures were still rising before the pause. Oh! Now they’ve discovered it!
e) unfortunately, the large heat capacity means that the putative temperature changes of "global-warming" would be very very small and so
f) ...this requires very, very sensitive detection instrumentation
g) ...over huge areas of the worlds oceans (in the near-surface areas)
h) ...over a significant period of time. The ARGO buoy system is a valiantly expensive attempt to start this, but has not run for long enough yet. But then
i)... the alarmist global-warming activists found that the initial ARGO buoy results were not even producing the desired warming expected, so started "adjusting" the initial ARGO results:
j)...meaning we will simply have to wait commensurately more years before the results start to become meaningful.
k) ...those adjustments to ocean heat content didn't come soon enough to get "observed" warming back up to where the models predicted it should be (Trenberth's "Travesty"), so we got
l) Trenberth suggesting that the recalcitrant missing heat may be hanging out in the deep deep ocean where, conveniently, it
m) cannot be measured by ARGO and
n) wouldn't produce any sea level rise even if it was in such places, which, until I can be bothered to think of yet more reasons, leaves it up to you to explain to Raff why he is
o) in the absence of further evidence a
p) enis.

Feb 27, 2016 at 11:03 PM | Unregistered Commentermichael hart

Martin A

I think of the CO2 increase and the resulting energy imbalance as the cause, global warming as the effect and climate change as the consequences.

In the context of OHC you can look at the increase as a crosscheck on the energy imbalance which causes global warming.

Alternately it can be seen as a cause of sea level rise, a form of climate change.

I read some of the comments at Climate Etc. It was disconcerting, but perhaps not surprising, to see respectable sceptic scientists being attacked by their own side for going off message.

Radical Rodent

The OHC might be increasing by this scary-looking figure, but how would this be, if expressed as a temperature? How many zeroes would be between the number and the decimal point, and on which side? 0.1°C (or 0.1K, if you prefer) does not look as scary, does it?

Perhaps you lack imagination.

Large numbers, at least by human standards, are routine on a planetary scale. The volume of the oceans is 1.3 billion cubic kilometres. Warming that volume even by 0.01C requires a lot of heat.

When you look at the post-1980 changes in the ocean all these statements are equivalent. They describe the effect of the same change in OHC in different units. Some sound large, some sound small.

Ocean heat content increase 20*10^22Joules

Rate of increase. 0.6*10^22Joules/year

Average temperature rise 0.06C

Sea level rise due to thermal expansion 49mm

Rate of sea level rise due to TE. 1.4mm/year

Feb 27, 2016 at 11:59 PM | Unregistered CommenterEntropic man

Martin A

EM - I see what you are saying (I think). That the effect of additional atmospheric greenhouse gases is to cause the ocean to absorb additional heat and so the best way to assess the impact of additional GHG's is by assessing OHC. Is that roughly it?

That's right.

Increased deep ocean OHC does not much affect climate, but it does give a measure of the imbalance and the long term rate of energy accumulation.

What does affect climate is that the imbalance increases the amount of energy passing through the surface and atmosphere. This extra energy remains at the surface for a while and increases surface temperatures in the shorter term. The higher the rate of flow, the higher the temperature change.

The extra energy eventually finds its way into the bulk ocean and accumulates as OHC.

Feb 28, 2016 at 12:48 PM | Unregistered CommenterEntropic man

Large numbers, at least by human standards, are routine on a planetary scale.
Yay! So, are you truly beginning to understand, or are you just patronising me? Alas, your previous sentence suggests the latter; sorry, big boy, but patronising will not work on me.

Now, how do you arrive at those figures, given the paltry amount of measurements that have been, and still are being undertaken in the oceans? (A question, I have noted, that you have assiduously avoided answering so far; this could be interesting.)

Feb 28, 2016 at 5:34 PM | Registered CommenterRadical Rodent

To repeat a previous comment, when EM mentioned that the heat accumulated annually in the ocean is a mere ten times* the annual human consumption of energy, I was surprised how small it is.

_______________________________________________________________________________

* To be clear, the word "mere" is mine, not EM's. I had the impression that EM felt that readers would be impressed that it was a really huge quantity.

Feb 28, 2016 at 6:30 PM | Unregistered CommenterMartin A

One more thing, EM: you state that the ocean heat content has increased 20*10^22 Joules since… when? You do not make that too clear; is my assumption that this is since 1980 correct? If that is the case, how was the ocean heat content calculated in 1980? Also, what was the rate of increase pre-1980? If we do not know the answer to that question, then how can we determine the relevance of the increase quoted? (Note that I have not bothered asking how this possible pre-1980 increase was determined; let’s keep it simple.)

Finally: 0.06°C? Are you serious? This can be measured? Presumably, you have calculated the rise due to thermal expansion by… how? Should you have used the ASTM tables, you should be aware that they are usually calculated to within 0.25°C or even 0.5°C (though some are still calculated in Fahrenheit), and interpolation is discouraged, as the differences are not always linear, ergo, a reading of 14.5°C will give the same volume correction factor as a reading of 14.56°C. (You might have access to one of the many ASTM formulae now extant, but even these should be calculated within the parameters of the tables, rounding to the nearest 0.25 or 0.5 degree, with similar restrictions on the density, SG or API used – a mistake made by rather too many others, who seem to think that having a formula and a spreadsheet gives them license to take things to stupid levels of “accuracy”, I am afraid.) You also need to know the density, API or SG for these; presumably you divined an average for your calculations, which, of course, means that you assumed that all different densities are increasing heat content at the same rate. While I can agree that assumptions have to be made in science, you can only use them as the basis for further research, not for final conclusions.

Feb 28, 2016 at 7:14 PM | Registered CommenterRadical Rodent

Radical Rodent

You are convinced that 5000 ARGO floats, plus buoys, is too small a sample size. Why?

Did you try any of the statistical method for relating sample size and confidence?

There is a useful calculator here.

For reference, an ARGO float generates a 200 data point temperature profile every ten days. Annually that is 3500*200*36.5=2555000 data points. Sample size for an annual average ocean temperature, n=25 million.

Feb 28, 2016 at 8:00 PM | Unregistered CommenterEntropic man

Radical Rodent

For published OHC content I used this NASA graph. . As you say, I used 1980 as a start point.

As we discussed once, the precision of a mean improves with increasing sample size according to the formula

Precision of the mean=precision of measurement/√sample size

If you assume that, very conservatively, each float measures to the nearest 1C the precision becomes 1/√25*10^7=0.0002C

Thus a change in mean temperature of 0.06C is well within the measurement resolution of the ARGO system.

For my own calculation relating OHC and sea level rise I used the thermal expansion coefficient for seawater at 700m depth and pressure..

Feb 28, 2016 at 8:25 PM | Unregistered CommenterEntropic man

Yes, I do, Entropic man. They are a good start, but you cannot seriously think that they have provided enough information, yet, to derive any reasonable conclusion. If n=25 million, that is still one reading per 44 cubic kilometres; however, n actually = 2.5 million, in the calculation you gave. How deep do the ARGO buoys go? 700 or 2,000 metres? What is the average depth of the oceans? Oh, about 4,000 metres – you are measuring less than half the oceans, here! Furthermore, how accurate are the instruments on the ARGO floats? How often are they checked and calibrated? There is still an awful lot of work to be done before we can start formulating realistic pictures of what might be going on. Have patience, and do not leap to conclusions quite so quickly. The planet is really a very big place, EM, so big that perhaps your imagination cannot quite grasp the enormity of it, and the pitifulness that humans are against it.

For my own calculation relating OHC and sea level rise I used the thermal expansion coefficient for seawater at 700m depth and pressure.
Please explain how; it is no good just giving an answer without explaining how you reached it, dear boy.

Feb 28, 2016 at 8:40 PM | Registered CommenterRadical Rodent

EM - Recently, you wrote

Please do not refer back to the Essex paper. I passed it on to an engineer acquaintance with more maths than myself for an independant opinion. His comment on its validity was negative.
Feb 25, 2016 at 7:25 PM | Unregistered CommenterEntropic man

We are still waiting with bated breath to learn what aspects of the paper by Essex et al caused your acquaintance to say that it was "invalid".

Please reveal this information as soon as possible; as I said before, it would be really interesting to hear a valid argument on that subject.

If nothing is forthcoming, we shall be obliged to conclude that the discussion with your acquaintance was something that occurred only in your imagination and, despite its purely imaginary nature, it became, for you, an event that happened in reality.

Feb 29, 2016 at 7:16 AM | Unregistered CommenterMartin A

EM

"Precision of the mean=precision of measurement/√sample size"

Not correct. This is only valid until you reach the resolution of the instrument, especially electronic ones. Any true value within the resolution would be rounded so you can't apply this theory. In its simplest sense it's the issue of measuring a human hair with a ruler that is marked out in mm. It doesn't matter how many measurements you take the uncertainty is still half a mm.

There is a technique called Gaussian blurring but in general stating precision below the resolution is fraught with difficulties.

Feb 29, 2016 at 8:00 AM | Registered CommenterMicky H Corbett

For reference, an ARGO float generates a 200 data point temperature profile every ten days. Annually that is 3500*200*36.5=2555000 data points. Sample size for an annual average ocean temperature, n=25 million.
Feb 28, 2016 at 8:00 PM | Unregistered CommenterEntropic man


Precision of the mean=precision of measurement/√sample size

If you assume that, very conservatively, each float measures to the nearest 1C the precision becomes 1/√25*10^7=0.0002C

Thus a change in mean temperature of 0.06C is well within the measurement resolution of the ARGO system.

Feb 28, 2016 at 8:25 PM | Unregistered CommenterEntropic man

Sorry EM, that's wrong. You are blindly applying what you memorised in the "statistics 101" class that you boasted about not so long ago. Mickey H Corbett is right, but it is worse than what he correctly points out.

Your √n formula depends on assumptions such as:

- the errors are solely due to the instrument
- the errors are random, independent, and identically distributed


You are trying to estimate m, the mean ocean temperature. An individual measurement i will give you a value
xi = ( m + delta i + epsilon i) where

m = the mean ocean temperature you hope to estimate.

delta i = the difference between the mean temperature m and the local temperature in the vicinity of the float at the ith measurement.

epsilon i = the instrument error at the ith measurement

If the float is somewhere warrm - eg in the Gulf of Mexico - delta will be positive and will remain so from one measurement to the next. If the float is somewhere cold - eg within the arctic circle - delta will be negative and remain so. So the delta values are not independent from measurement to measurement. Nor is there any reason to think they will be identically distributed. Moreover, they will generally be very much bigger than the error in the float's temperature sensor.

Your √n formula does not apply and the assumption that (x₁ + x₂+ ... + xₙ)/n will give you an estimate of m to a precision ± epsilon / √n is just plain wrong.

Another for the collection.

Feb 29, 2016 at 9:48 AM | Registered CommenterMartin A