Continuation of the older threads. Please scan those (even briefly) to see whether your point has already been dealt with. Let me know if there is something worth pulling from the comments to the main post.
In the meantime, read about why peer-review is a necessary but not sufficient condition for science to be worth looking at. Also, before you conclude that the emails have any impact on the science, read about the six easy steps that mean that CO2 (and the other greenhouse gases) are indeed likely to be a problem, and think specifically how anything in the emails affect them.
Update: The piece by Peter Kelemen at Columbia in Popular Mechanics is quite sensible, even if I don’t agree in all particulars.
Further update: Nature’s editorial.
Further, further update: Ben Santer’s mail (click on quoted text), the Mike Hulme op-ed, and Kevin Trenberth.
Scott A. Mandia (037), interesting but somewhat confusing: 1) Did Mt Pinatubo warm or cool the stratosphere? 2) Is ozone increasing or decreasing in the periods of your post? You say, “… This is no surprise because ozone levels are increasing which should be causing a warming trend.” and then, “…Model calculations suggest that the upper stratosphere trends are due… to decreases in ozone….” Or am I just not following you correctly?
Max, you remind me of the guy who shoots his arrow at the barn, then paints concentric circles around it.
You don’t make logical leaps — you state what you believe, over and over, and the rest is just throwing words around.
Like picking six out of 160 data points to create a line on a page, then claiming it proves what you wanted it to.
Max(1042)
Thanks. I now realized you are talking about the offset since 2000 and some other forcing given by something other than CO2, not denying the AGW theory. I have no objection on that view.
best
Max is j*ing himself off on the “oh yeah, I know about CO2, but how do you know that ‘something else’ isn’t going on?” riff.
This guy’s car overheats. Mechanic tells him the water pump is shot, going to cost him bucks. Second opinion; it’s the water pump. Third opinion; the same. Etc. Finally his neighbor (no mechanic) tells him, don’t worry, could be the temperature guage. Sure enough, the next day is cold, car doesn’t overheat. A week later the engine blows. BIG bucks.
Another lousey analogy, but it works for me.
Here’s what I don’t understand…..if it’s warming how come some US state are breaking records for lowest recorded winter temps and record snow falls.
Anyone care to explain this to me in laymans terms?
Rod B #1050:
Sorry for the confusion:
1) Pinatubo warmed the lower stratosphere. The downward steps therefore occur as this short-term warming ended.
2) There are two periods of discussion. Randel et al looked at the entire stratosphere between 1979 and 2005. That period showed a cooling trend in the entire stratosphere but the largest cooling was in the upper stratosphere.
However, in the lower stratosphere the cooling trend stopped in 1995. This is just about the same time that the ozone in the lower stratosphere began recovery. It appears that ozone depletion was the primary cause for the decline in lower stratosphere temperatures.
The upper stratosphere cooling over that time frame was estimated to be about 50/50 between ozone loss and increased CO2 for the entire time frame.
Does this make sense? I have to admit I am having trouble wording this.
Comment by cold in the US — 17 December 2009 @ 9:30 PM
“Here’s what I don’t understand…..if it’s warming how come…”
Snowfall is governed by temperature and availability of moisture, among other things. To the extent that air is cool enough to precipitate out moisture as snow, record snowfall is actually quite consistent with a warming planet. More evaporation, more moisture available in the air, more moisture thus available to precipitate as snow in those places with sufficiently low temperatures.
There’s also nothing particularly weird about seeing record lows established in particular places even as temperatures in general head upward. That’s why it would be crazy to make a judgment about trends in global temperatures from a single thermometer.
As to records in the U.S., you might be interested in this, though any real dyed-in-the-wool contrarian will simply scoff while dropping down the “all data is bad” rabbit hole:
http://scienceblogs.com/islandofdoubt/2009/11/record_high_and_low_temps_an_i.php
Gist is, more record highs are being established than record lows, which is exactly what we expect.
Rod B wrote in 1050:
I don’t know about confusing. To me it is a bit more like “detailed.” There is the lower stratosphere, the upper stratosphere, the stratosphere above 35-40 km, and there is the period prior to 1995 and the period after 1995. He is trying to give you the lay of the land… or I suppose “the lay of the stratosphere” would be more like it. But as I said, somewhat detailed but not really confusing.
Perhaps I can help…
Rod B wrote in 1050:
Scott said, “… warming associated with the El Chichon and Pinatubo volcanic eruptions.”
Volcanos warm the stratosphere, so to filter out their effects Randel et al (2009) omit the two years of temperature records immediately following a major eruption (i.e., El Chichon in 1982 and Mt. Pinatubo 1991). So Mt. Pinatubo warmed the statosphere, briefly.
Rod B wrote in 1050:
The period that Randel et al (2009) are concerned with is from 1979 to 2007. (Looking at history legislation against the use of CFCs only started being put in place by 1980 and it wasn’t all at once. Moreover, CFCs tend to stay in the stratosphere for a very long time.) Now Scott specifically states that, “the lower stratosphere has not noticeably cooled since 1995. This is no surprise because ozone levels are increasing which should be causing a warming trend.” So it would appear that ozone in the lower stratosphere was beginning to recover by about 1995.
So ozone would appear to have been decreasing prior to 1995 but leveling off and then recovering after 1995.
Rod B wrote in 1050:
That would be after 1995 — and mostly the lower stratosphere.
*
Rod B wrote in 1050:
I believe what Scott said at this point was:
So even if ozone begins to recover in the upper stratosphere given the increase in well-mixed greenhouse gases (if I remember right, carbon dioxide and water vapor, mostly), it is no surprise that the upper stratosphere continues to cool longer than the lower stratosphere, that is, beyond 1995. Furthermore, earlier Scott states, “In the middle and upper stratosphere there was mean cooling of 0.5–1.5 K/decade during 1979–2005, with the greatest cooling in the upper stratosphere near 40–50 km. Ozone concentration above 35 km is minimal so ozone depletion is much less a factor at these levels than cooling due to CO2.” So it would appear that the cooling trend continues above 35 km — pretty much independently of what is happening to the ozone.
Rod B continued:
Quite a bit those three little dots are standing in for, don’t you think? But at least you remembered to put those dots in though, I suppose.
Anyway, in the upper stratosphere but below 35 km, cooling was about equally due to ozone depletion and growing levels of carbon dioxide. But since ozone levels are beginning to recover, in the lower stratosphere you are seeing the cooling level off as of 1995 since the effect of ozone is to warm the stratosphere. Above that the cooling has moderated, and above 35-40 km ozone has a negligible effect so the cooling is continuing pretty much as it had before.
It helps to remember that the mechanism by which ozone warms the stratosphere is essentially the same as greenhouse gases — ozone is able to absorb certain bands in the infrared and performs the role of a greenhouse gas in the troposphere. But it is also able to absorb certain bands in the ultraviolet and as such warms the lower stratosphere with energy that comes directly from sunlight rather than simply the thermal radiation from below.
*
Rod, just two years ago you were really getting into the quantized states of molecular excitation that are responsible absorption and emission of infrared radiation by greenhouse gases, the different modes of thermal excitation, even the non-local thermodynamic equilibria, and now… I do hope things get better.
PS
The message I was quoting from Rod B. is now at 1051.
Hank Roberts (1052)
Hank, you’re rambling.
What, exactly, is your point?
Max
1060 manacker says: 18 December 2009 at 3:38 AM
Hank Roberts (1052)
Hank, you’re rambling.
What, exactly, is your point?
……………………………………………
Manacker, it’s as clear as day. Given your confusion about when the 21st Century started (we don’t use the Julian calendar these days), I suggest you read his comment again… and again… and again… It may eventually sink in.
RE manacker
That maybe you’ve already decided on a conclusion and are picking only the information that supports said conclusion? And continue to restate that conclusion and the limited set of facts? I dunno, I could be wrong.
Max “manacker” pretends ignorance. People have pointed out he’s faking his charts:
http://www.google.com/search?q=site%3Arealclimate.org+manacker+six+160
P.S., the prolific “manacker” at Salon, for those who don’t know how much of this material he produces. You can find the same stuff in volume at many other sites discussing climate:
https://letters.salon.com/ace592d12ba8a39436beb337e351dcc4/author/index2.html
Thanks, Chris, for stating this much better than I. :)
I will try to re-word my site to more closely to match your explanation. (And if so, I will give you credit of course.)
Agree with Gerson? Yes this part right here:
“But the hacked climate e-mails reveal a scandal, not a hoax. Even if every question raised in these e-mails were conceded, the cumulative case for global climate disruption would be strong. The evidence is found not only in East Anglian computers but also in changing crop zones, declining species, melting ice sheets and glaciers, thinning sea ice and rising sea levels. No other scientific theory explains these changes as well as global warming related to the rise in greenhouse gas emissions since the Industrial Revolution. Over millennia, the climate shifts in natural cycles. But we seem to be increasing the pace of change so rapidly that plants, animals and humans may not be able to adequately adjust.
The claim of recent global cooling is deceptive.”
(Apparently I can’t let well enough alone)
For everyone interested:
How to make a significant linear trend non-significant:
http://img51.imageshack.us/img51/65/crazysim.jpg
This data is _simulated_ by adding gaussian noise to f(x) = sin(x)+exp(x/15). The red line is fit to all data, the blue line only to the highlighted blue points. Blue points in residual versus x plot in bottom panel as well.
Note also that _outliers_ is a technical statistical term and is only defined relative to a given model. The picking the peaks and troughs of the cycle in this case is tantamount to fitting outliers. A fit only to those points will be less significant than a fit to all the data.
For those interested, this R code produced the pair of graphs I just posted:
http://img32.imageshack.us/img32/7734/picture5oc.png
(posted as a screen capture b/c the actual code gets flagged as spam…)
Sorry, Hank, you are as wrong about when the 21st century started (1061) as you are about most of your opinions.:
See:
http://www.mayorscommission.org/faq/files/sc21cent.cfm
From Wiki:
http://en.wikipedia.org/wiki/21st_century
Max
[Response: So no parties for you on new year’s eve 1999 then? – gavin]
Gavin
Yeah. We had a “Y2K party” (was in Hong Kong). Observed (with great relief) that the world did not end due to massive computer crashes.
But it was not a “new millenium” party.
Max
> 1061
Try reading before writing, Max, you could develop attribution skills, and at least start to insult people intentionally instead of accidentally.
Still awaiting a _rational_ response:
“Can you explain how you calculated your temperature anomaly? Can you explain why they are so far off from the time series from which you claim to have sourced your data? Seeing as so much data is available from your source, why did you winnow 160 anomaly data points down to 6?”
Comment by jay — 14 December 2009 @ 6:50 PM
Doug Bostrom
The Meehl et al. study on trends in U.S. high/low extreme temperatures, which you cited (1057) is interesting, but Meehl has only started his record in 1950, thereby cutting off the period with the most extreme high temperatures, which occurred prior to 1950.
Taking a look at the record high/low temperature by states and extending the record to cover record temperatures registered before 1950 gives a different picture.
http://en.wikipedia.org/wiki/U.S._state_temperature_extremes
1880s: 1 high, 0 low
1890s: 1 high, 5 low
1900s: 1 high, 4 low
1910s: 5 high, 4 low
1920s: 2 high, 0 low
1930s: 23 high, 12 low
1940s: 0 high, 2 low
Subtotal: 33 high, 27 low (prior to 1950); ratio 33/27 = 1.22
1950s: 5 high, 2 low
1960s: 1 high, 5 low
1970s: 2 high, 4 low
1980s: 2 high, 6 low
1990s: 5 high, 5 low
2000s: 2 high, 1 low
Subtotal: 17 high, 23 low (after 1950); ratio 17/23 = 0.74
[Note that Wiki lists the latest date a record was reached, in case of a tie.].
1920s/1930s: 25 high 12 low; ratio 25/12 = 2.08
1990s/2000s: 7 high, 6 low; ratio 7/6 = 1.17
Just shows you can prove anything by picking your starting point properly.
Before the howls of outrage start, Doug, I do not make the claim that these numbers show any kind of trend. They just cover extreme temperatures by US state measured prior to Meehl’s arbitrary 1950 cut-off.
They do show that the 1920s/1930s had more extreme highs (and more extreme lows) than the 1990s/2000s, also with a higher ratio of highs to lows.
So the conclusion that the ratio of record highs to record lows has increased in the 1990s/2000s over earlier periods does not hold, if the time period prior to 1950 is also considered.
Max
Hank Roberts (1063)
You apparently did not like (or maybe were unable to understand) the chart I posted showing the change in atmospheric CO2 versus change in temperature for various observed multi-decadal oscillations in the HadCRUT temperature record.
You can see these oscillations more clearly, along with the linear trend for each multi-decadal period, on this chart of the complete HadCRUT record.
http://farm3.static.flickr.com/2545/4193198611_d7e7d2f1e4_b.jpg
No “faking”, Hank – just the facts.
Max
[Response: Matching linear trends are not ‘facts’ and any random application of a low-frequency smooth and trends through the residuals could be used to get anything you want. This kind of ‘analysis’ is completely arbitrary and proves nothing to anybody. – gavin]
Gavin (1073)
There has been no “smoothing”, Gavin. The record as published shows multi-decadal linear warming and cooling trends, as indicated, pointing out that the observed temperature record has gone through oscillations, averaging about 60 years for a complete cycle.
What this “proves” is anyone’s guess, but it is not “insignificant” just because it is “inconvenient”.
The early 20th century warming period (1910-1944) has been cited (and studied by Delworth and Knutson); the late 20th century warming period (starting around 1976) has been covered extensively by IPCC; the mid-century cooling has also been cited by IPCC.
There has been less attention given to the earlier warming and cooling cycles.
So the cycles are recognized, even if they cannot be fully explained.
Max
[Response: Look, what you are doing is attempting to define something as the long term (presumably forced) trend and then look at the residual variability to characterise it. You have arbitrarily used a linear regression to define the long term component, while you could have used any number of things – a quadratic, a cubic spline, a low-frequency band pass, wavelets, EMD, the AR4 ensemble mean etc. From any of these curves you could then define a residual which you would like to call a ‘cycle’ of some sort. However, the cycles you find are completely determined by what you arbitrarily defined as the trend. But, and here’s the problem, the forced component of climate change isn’t forced to follow any arbitrary smoothing function (linear or otherwise). To declare that your particular decomposition is a ‘fact’ is on a par with declaring that the Cydonian hills contain a face. It is a complete artefact. Is their multi-decadal internal variability? sure. But residuals from linear trends do not define it. – gavin]
Somebody said “So the conclusion that the ratio of record highs to record lows has increased in the 1990s/2000s over earlier periods does not hold, if the time period prior to 1950 is also considered.”
Odd, Meehl’s paper attempted no such conclusion. How could somebody infer such a claim? Weird.
Gavin
I appreciate your knowledge of statistics, but I believe it is always best to look at the most simple, straightforward relationships first.
The unadulterated HadCRUT temperature record shows that there have been several multi-decadal warming and cooling cycles over the 159-year record.
These cycles are not, as you say “completely determined by what I arbitrarily defined as the trend”.
Instead, they are easily recognizable in the overall record. Dr. Akasofu has also identified them.
They are real, Gavin, even though we cannot fully explain them yet. I am hopeful that climate science will some day be able to explain them, but we are not there yet.
Their linear warming or cooling rates can easily be identified without any “arbitrary smoothing”.
The record is what it is, Gavin. If you prefer not to call it a “fact”, that’s OK, but it is the observed record.
Max
Max’s “calculations” in #1072 could be realised with this dataset (snippet produced because I can’t be arsed to type in much more ‘cos that barnpot isn’t going to notice):
1
2
1.5
2.5
2
4
3.75
3.5
3
4.5
3.5
5
.
.
.
Now what’s the trend?
5 up, 6 down, therefore going down?
Max: I think you need to play with some fake data of your own. Use Excel’s random number generation function: take an underlying trend over time (say, the log of the CO2 concentration): add, say, a small 11 year sin function to represent solar cycles: add maybe another 4 year function times a random number to represent ENSO (though it isn’t strictly four years): add an “aerosol cooling” trend that rises in the early period and falls in the late period: and add another random number residual which has some auto-correlation (eg, the random number in year 7 is somewhat dependent on the number in year 6).
Then generate a few outputs and graph them over time. You’re going to get things that look like “cycles” of uneven time periods – 30 years here, 20 years there – and you’ll be able to divide this time period up and assign up trends and down trends. But they won’t _mean_ anything: by definition, the only trends that mean anything are the ones defined by the algorithm.
_This_ is what attribution is all about: taking a noisy signal and trying to figure out what it comes from. But it requires some physical understanding of the system – you can’t get anywhere just doing random statistical manipulations. (and in real attribution there is a lot more data: hemispheric patterns, continental patterns, ocean heat patterns, stratospheric temperature trends, that allows to be more confidence about conclusions)
Still awaiting a rational response from Max to:
“Seeing as so much data is available from your source, why did you winnow 160 anomaly data points down to 6?”
Comment by jay — 14 December 2009 @ 6:50 PM
Hank Roberts (1079)
To answer jay’s loaded question, Hank. I did not. Sorry.
I just analyzed the record as it has been published.
Please refer to the graph I posted for clarification (1073) and stop parroting the same question over and over again.
Max
manacker wrote in 1073:
Strictly speaking, the “oscillations” you’ve identified aren’t periodic but quasi-periodic. Some as short as eight years, others longer than thirty. And consider: with any noise that that is left over from something roughly tracking a linear trend (e.g., where the linear trend is the straight line you have going from 1850 to present) the noise will result in a jagged little curve that goes below the trend for a while, descending to its lowest point below the linear trend, then goes up for a while until it is above the trend before it starts descending for a while, following the same “pattern.”
Now of course there is a “mathematical structure” to the temperature record, something which I think people can pick up on visually by looking at the temperature record — as the result of our aesthetic sense. It is called 1/f noise, pink noise or flicker noise. It exhibits a fractal self-similarity. But it isn’t specific to the temperature record. It is a very common phenomena of the natural world. You will find it in the sound of the rain and the blowing of the wind, the stops and goes of traffic past a given point, the waves of people going that come to and leave a restaurant. Music imitates this mathematical structure, exhibiting 1/f noise that permeates our stop-and-go world of change.
Then of course there are the actual climate oscillations. El Niño Southern Oscillation, the Pacific Decadal Oscillation, Artic Oscillation and so on. They aren’t periodic, but they are quasi-periodic — and we understand these to varying degrees.
We are able to explain the role of the trade winds, Walker circulation and Kelvin waves in quasi-periodically causing the thermocline to alternate between upwelling in the western then eastern Pacific.
For a somewhat detailed explanation you might try:
http://www.ldeo.columbia.edu/edu/dees/V1003/lectures/ENSO/ENSO.2100.pdf
However, climate oscillations are neither able to create nor destroy energy. They simply move it about. And if the atmsophere becomes more opaque to thermal radiation given the absorption specta of greenhouse gases and greater amounts of those gases in our atmosphere, the rate at which the climate system radiates away thermal energy will necessarily decline. Then given the principle of the conservation of energy, the climate system will warm up until the temperature is high enough that it compensates for the increased opacity of the atmosphere and the earth radiates thermal radiation into space at the same rate as before.
Climate oscillations don’t change this. They simply move thermal energy from the surface of the ocean to the ocean depths, from the lands surface to the higher layers of the atmosphere, or between the atmosphere and the ocean — then back again. And the movement of heat within the system results in much of the “natural variability” we see in the temperature record. But it does not change the underlying physics. It does not change the principles of quantum mechanics, their application to the quantized states of molecular excitation in gases, liquids or solids, the absorption spectra of greenhouse gases, or any of the basic principles of physics. They simply redistribute the heat — heat that escapes the climate system only as thermal radiation which must make it through an atmosphere that is partly opaque to it.
Completely Fed Up (1077)
Playing with imaginary numbers may be fun for you, CFU, but I prefer real data, as I pointed out to Marcus.
Max
Timothy Chase (1080)
You wrote:
Wrong, Timothy. Look more closely.
The shortest is 22 years (19th century); the longest 35 years. The 8-year periods at the beginning and end of the data series are clearly identified: “short term periods may or my not be part of a multi-decadal cycle”.
The overall cycle of the oscillation appears to be somewhere around 60 years.
Max
> Max/manacker:
> I just analyzed the record as it has been published.
No, you analyzed six points you picked out of the published record.
To analyze the record as published, you’d use the data as published.
This isn’t even blog science, Max, it’s assertion by bluster.
Timothy Chase says: 19 December 2009 at 1:47 PM
Last two paragraphs really excellent, in particular this can serve as a touchstone whenever mention is made of various periodic phenomena:
“However, climate oscillations are neither able to create nor destroy energy. They simply move it about. They simply move it about. And if the atmosphere becomes more opaque to thermal radiation given the absorption spectra of greenhouse gases and greater amounts of those gases in our atmosphere, the rate at which the climate system radiates away thermal energy will necessarily decline. Then given the principle of the conservation of energy, the climate system will warm up until the temperature is high enough that it compensates for the increased opacity of the atmosphere and the earth radiates thermal radiation into space at the same rate as before.”
Rearranging deckchairs to solve the problem of too many chairs arriving on deck does nothing to address oversupply. Fixing the problem of crowded decks requires throwing chairs overboard fast enough to avoid being crushed under a pile of wood and canvas. If the ship’s rails are too high, chairs can’t be jettisoned fast enough.
Distilled to the important essence this is not very complicated.
Timothy
You stated (1081):
How true. Only the sun (including any climate oscillations that may be caused by the changes in the incoming energy from the sun) can do that.
Clouds and may reflect incoming SW radiation or absorb and reradiate outgoing LW radiation , GHGs may absorb and reradiate outgoing LW radiation, surface albedo may reflect incoming SW radiation, etc.
But the sun is the only significant source of energy our planet has.
Max
manacker (Max Anacker), you have brought up the “positive feedbacks” before as something that you seem to think is grafted on to the climate models. By itself a doubling of carbon dioxide results in an increase in temperature of 1.1-1.2°C, but then the feedbacks of the climate system result in an additional 1.6°C. That sort of thing.
However, climate models themselves do not calculate things in terms of forcings or feedbacks. Rather, that is how we conceptualize the response of the models or the response of the climate system that they model to increased levels of solar radiation or greenhouse gases in the atmosphere. And this is something that is done after the fact — as a way of summarizing how the systems respond.
But the actual climate models are based upon well-established principles of physics. They don’t rely upon mere observed correlations between carbon dioxide and temperature. They don’t rely upon arbitrary grafts.
If you raise the temperature of the climate system by increasing the opacity of the atmosphere to infrared radiation (something we are able to see with infrared imaging both at the surface and from satellites), then the rate of evaporation will increase. For every degree Celsius you increase the temperature the absolute humidity will increase roughly 8%. For every ten degrees Celsius the absolute humidity will roughly double. And in accordance with its well-established absorption spectra, water vapor — just like carbon dioxide — will decrease the rate at which thermal energy can be radiated into space.
When the rate at which energy enters the system remains the same but the rate at which energy leaves the system is diminished the level of energy will necessarily increase. And it is only as the system heats up that the rate at which thermal energy is radiated compensates for the increased opacity of the atmosphere such that the temperature levels off at a new, higher equilibrium.
The amplification is due to well-established principles of physics, not some arbitrarily hypothesized climate sensitivity. “Climate sensitivity” — as climatologists employ the concept — is simply a way of summarizing for convenience — how the system responds — not for calculating how it will respond.
Really off topic, but regarding post 1072 et al, have those here with the capacity donated to Wikipedia?
I did, and I’m positively glowing with virtuous smugness.
Back on topic (roughly speaking), the spurious claim about Meehl’s conclusions at 1072 has a lot to do with overlooking the important takeaway of the paper in question. It’s really not about the number of record highs versus the number of record lows per se, but the relationship between observed record highs and lows and how they comport with model predictions. Meehl’s work reinforces the predictions of models regarding that relationship. Record lows are showing a slowing rate of excursion more than record highs. That’s why the simplistic analysis in 1072 really is not even aimed at the side of the barn in question.
While looking for Meehl’s paper on the web I notice that 1072’s blunder has been repeated and even proudly touted in many places. Pretty embarrassing.
Closing Press Briefing on United Nations Framework Convention on Climate Change.
http://unfccc.int/2860.php
Does anyone have any comments regarding the 2 degree C commitment?
Is this achievable?
Is it relevant?
[Gavin. Hope this is not too far OT.]
Max
Timothy Chase
An interesting dissertation on feedbacks, models, etc. (1087)
Thanks.
I think a shorter summary is to be read on p.12 of the IPCC 2007 SPM report:
Max
Congratulations to Max on making the short list, albeit the very last guy:
http://www.joabbess.com/2009/12/02/show-us-your-emails/
manacker wrote in 1086:
Kirchoff’s law, Max.
Derived from the second law of thermodynamics, it shows that the spectral absorptivity of a substance or gas in local thermodynamic equilibrium with the surrounding electromagnetic field is equal to the spectral emissivity. A good radiator is also a good absorber. Clouds radiate thermal radiation, but they also absorb thermal radiation and have associated with them a greenhouse effect of their own. And at this point the studies that are coming in show them to be a positive feedback, not a negative one.
*
manacker wrote in 1086:
Max, what matters in terms of the average temperature of the climate system isn’t simply the rate at which thermal energy enters the climate system, but the rate at which thermal energy escapes the climate system. But for solar cycles solar radiation entering the climate system has been flat to slightly dropping since at least 1960 and more likely 1950. This in itself would result in a downward trend in global temperature. Not the rising trend that we saw from about 1975 forward.
The level of greenhouse gases in the atmosphere, however, has risen, increasing the opacity of the atmosphere to thermal radiation. This is something that we can understand in terms of the fundamental principles of quantum mechanics, the most accurate and well-established theory in physics today. It is something that we can measure at the earth’s surface. It is something that we can observe by means of satellite.
It isn’t the rate at which energy enters the system that has changed. It is the rate at which energy leaves the system. Furthermore, if it were increased solar radiance then we would have seen the stratosphere warm with the troposphere rather than cool as the surface warms. But it cooled. If it were increased solar radiance then we would have seen the days warm more quickly than the nights. But it is the nights that warmed more quickly.
All evidence points towards an enhanced greenhouse effect, not increased solar radiance. And a conclusion supported by multiple independent lines of evidence receives far more justification than it would from any one line of evidence considered in isolation from the rest. It is our acknowledgement of this principle that is the reason for why science works, Max.
Timothy Chase
We may be drifting too far OT here, but let’s see if Gavin cuts this exchange off.
You wrote (1087):
IPCC (Myhre et al.) tells us that a doubling of CO2 would cause a theoretical GH increase in temperature of just under 1°C.
IPCC (AR4 Ch 9) tells us:
With no feedback, 2xCO2 impact is predicted to be around 1°C.
With all feedbacks, excluding cloud feedback, it is predicted to be 1.9°C.
With all feedbacks, including cloud feedback, it is predicted to be 3.2°C (i.e. adding in cloud feedback increases 2xCO2 impact by 1.3°C on average).
All GCMs predict positive feedback from clouds, but disagree on magnitude.
IPCC tells us, “cloud feedbacks remain the largest source of uncertainty”.
Yet, despite this “uncertainty” the models all predict positive net feedback from clouds amounting to a 2xCO2 impact of 1.3°C out of the predicted 3.2°C.
Can you explain this?
Thanks.
Max
manacker wrote in 1083:
Max, I see in your table seven legs: 8 years (cooling), 22 years (warming), 31 years (cooling), 35 years (warming), 32 years (cooling), 25 years (warming), and then 8 years (cooling).
I do not see an oscillation of 8 years in length. That would be something repeating itself every 8 years. I do not see an oscillation of 16 years in length with a period of eight years of cooling followed by a period of eight years of warming — then repeating itself.
You show nothing of the sort. There are no oscillations of 8 years or 16 years in length.
60 years? I presume you mean 8 years of cooling followed by 22 years of warming followed by 31 years of cooling. I don’t see that repeated in your table. That would be with one cycle beginning with 8 years of cooling, ending with 31 years of cooling, immediately followed by 8 more years of cooling.
And at this point I have no idea how you would try and justify separating the 31 years of cooling at the end of one “cycle” from the 8 years of cooling at the beginning of the next. Except of course the “beginning” of the next “cycle” isn’t 8 years of cooling, but 31 years of warming.
What is it that you think is repeating every 60 years? “Around” 60 years? And assuming something existed that didn’t regulary repeat itself but only somewhat regulary, why would you consider such a thing to be periodic rather than quasi-periodic?
This doesn’t rise to the level of tea leaves.
Timothy Chase (1094)
Please read the footnote. [I will put this in bold face so it is easier to see.]
The two 8-year periods (at the beginning and end of the total time series) may or may not be part of a multi-decadal cycle, as noted.
Ignore them and concentrate on all the other oscillations. These show a total warming/cooling cycle of approximately 60 years each.
See it now?
It’s really not that complicated.
You ask what I think it is that repeats itself every 60 years.
I do not know.
All I can see from the record is that there are repetitive multi-decadal oscillations of approximately 60 years while atmospheric CO2 has not gone through such multi-decadal fluctuations, but has risen fairly steadily, accelerating after WWII.
So I ask you.
What do you think has caused these observed multi-decadal oscillations in the temperature record?
It it some sort of as yet unexplained natural variability or natural forcing?
Max
Could run FFT on GISEMP after the linear (rising) trend has been removed. Probably somebody has already done it.
Looking at that would save a lot of (silly) argumentation.
Hank Roberts
The annual HadCRUT record, which I plotted and analyzed has 159 points from 1850 to 2008 (not 6).
Repeating something wrong ten times does not make it come true, Hank.
Max
manacker wrote in 1093:
The figure I get is 1.1-1.2°C. Pretty much the same.
manacker wrote in 1093:
According to the text you quoted, 1.9°C±0.15°C.
manacker wrote in 1093:
With an uncertainty of ±0.7°C for the climate sensitivity — which I believe is larger than the uncertainty for forcing and all non-cloud related feedbacks as 0.7 is larger than 0.15. Yes, I believe I am following you so far.
manacker wrote in 1093:
Sounds about right.
manacker wrote in 1093:
1.3°C — with what range of error?
If one were entirely naive when it came to statistics one would assume that ranges of error are additive. Assuming this, given a range of error for everything without clouds of only 0.15 and a range of error for everything with clouds of 0.7, one would conclude that clouds add an uncertainty of 0.55. In which case this would certainly seem compatible with the claim that cloud feedbacks are the largest source of uncertainty.
But as it is, clouds could have associated with them a larger uncertainty than climate sensitivity itself. Given the law of large numbers uncertainties will often cancel one another out, and therefore — to consider a different sort of example — the uncertainty of the average temperature may be considerably smaller than the uncertainty associated with any individual temperature going into that average. And likewise the uncertainty associated with feedback due to clouds and the uncertainty associated with feedback due to aerosols may both be larger than the uncertainty associated with climate sensitivity.
Furthermore, climate sensitivity isn’t simply constrained by the behavior of the models but by the paleoclimate record. A metastudy:
Royer et al (23 Feb 2007) Climate sensitivity constrained by CO2 concentrations over the past 420 million years, Nature 446, 530-532
… based upon 47 different studies using 5 different methods concludes that the best fit climate sensitivity for the past 420 million years is roughly 2.8°C — which is the figure I typically go with. It is fairly close to what one gets from the models, albeit a little lower than the 3.2°C you quote. (Consistent though — as the range of uncertainty for each estimate overlaps with the other.)
Multiple independent lines of evidence, Max.
Manacker, your claim about a 60-year cycle is nothing more than seeing “pictures in the clouds.” You haven’t established it because there’s really no evidence for it.
This isn’t the first time you’ve claimed things that have no real basis in evidence. You do that all the time. It’s tiresome.
manacker wrote in 1095:
I am not speaking of a physical mechanism — I am just asking in terms of your numerology.
I said:
Leave the eight years in at the ends, take them out — whatever.
What sequence of numbers or set of tea leaves do you see being repeated?
manacker wrote in 1095:
CO2 isn’t the only forcing.
There is solar variability, methane, land-use, aerosols — and the clean air laws that started coming into effect during the 1970s and thereby reduced the latter. Temperatures rose during the 1930s, as one would expect with the reduction in industrial output reducing the levels of reflective aerosols — which in the troposphere has a half-life of only about 7-10 days.
Reduction in industrial output would have also reduced the rate at which carbon dioxide was being put into the atmosphere, but carbon dioxide has a much longer “half-life” with 25% remaining in the atmosphere after the first 1000 years — if I remember correctly. As such the effects of carbon dioxide emissions are cumulative — unlike aerosols. Ramping up for World War II would have increased sulfates once again, resulting in a downward trend in temperatures.
However, this would have happened principally in the northern hemisphere, not the southern as industrial production was primarily in the northern hemisphere and aerosols wouldn’t have stayed in the atmosphere long enough to affect the southern hemisphere as much. So we saw a statistically significant downward trend in global temperatures of only about 6 years, in the southern hemisphere of only 1 year, but in the northern hemisphere of about 30.
Clean air laws came into affect in the industrialized north during the seventies. This would have reduced the global dimming due to sulfates and unmasked the radiative forcing due to greenhouse gases. And naturally enough temperatures took off during the mid 1970s.