Guest essay by David Bennett Laing
In 1900, Knut Ångström concluded from a famous experiment that very little warming results from a doubling of atmospheric CO2. Although no similar experiment has been performed since, there is a simple and accurate way to tell by observing hard data from the real world. My approach in this is synthetic and not analytic, and I feel strongly that the integration of a variety of empirical data from the Earth system is essential for a good understanding of how things actually work in Nature.
Here in figure 1 is the Keeling curve, showing the steady increase in CO2 from 1959 to the present, as measured at the observatory on Mauna Loa, Hawaii:
Notice that there is an annual cycle (small squiggles) superimposed on the general increasing trend. If you average the monthly records for these cycles for each year, you wind up with the curve in the inset, which shows that, on the average, CO2 reaches a maximum concentration in the month of May in the northern hemisphere, when the CO2 from the winter’s decaying vegetation has warmed up and can enter the atmosphere, and a minimum in September-October, when photosynthesis during the summer has used up some of the CO2 in the atmosphere. The difference between the maxima and the minima is about 6 parts per million (ppm).
I have reproduced this inset curve for CO2, peaking in May, in the following graph as figure 2 (blue curve, all values in percent):
I took the NOAA record of northern hemisphere temperature anomalies (red curve, above) and gave it the same treatment for the 24-year period 1975 to 1998, when the globe warmed dramatically by nearly one degree centigrade. Notice that the temperature anomaly has its maximum value in March, two months before the maximum of the CO2 curve.
Now, if variations in the concentration of CO2 in the atmosphere had any effect on temperature, you would expect that the peak in CO2 would occur with or before the peak in temperature, but it actually occurs two months afterward, which shows that variations in atmospheric CO2 cannot possibly have a significant effect on temperature. Notice, however, that there is a very small up-tick in the temperature anomaly in June, showing that in fact CO2 actually does have a slight effect on temperature, but not a significant one.
What does the green curve represent? I gave the same treatment to the record of ozone depletion from Arosa Switzerland, for the same 1975 to 1998 period of rapid warming. Notice that it peaks in March, the same month in which temperature peaks. Both these peaks are sharp, suggesting a strong relationship between the two (the correlation coefficient between the red and green curves is 0.92, i.e., very high).
This means that there is a strong probability that it is ozone depletion, rather than CO2, that is really responsible for global warming. The interval 1975 to 1998 also happens to be the period during which chlorofluorocarbons (CFCs) were introduced to the atmosphere by spray cans and leaky refrigerators, as is shown by the graph below (effective chlorine means all anthropogenic sources of chorine and bromine):
As is well known, CFCs break down in March (in the northern hemisphere) on polar stratospheric clouds, and the chlorine and bromine thus released cause ozone depletion, including the ozone hole. The Montreal Protocol took effect in the ’90s and stopped the ongoing destruction of ozone, but because chlorine and bromine destroy ozone catalytically, most of these chemicals are still up in the stratosphere and are still destroying ozone, as the graph above shows. It will therefore take until at least mid-century for temperature to return to normal.
But why should ozone depletion cause global warming? That happens simply because the ozone layer shields Earth from Sun’s ultraviolet-B radiation (UV-B), which is 48 times hotter than Earth’s infrared radiation. When ozone is depleted, more UV-B can penetrate to Earth’s surface. In the late 20th century, however, everyone was concerned about this causing genetic damage and sunburn, but no one thought it could cause global warming as well. Apparently, it can, however, as is shown by the second graph.
All this very nicely explains why temperature rose so fast in the late 20th century, and why it leveled off again in what has been called the “global warming hiatus” (as if Earth were doing something wrong relative to the climate models!), but why did temperature jump up again in 2015?
This probably happened because from mid-2014 to early 2015, one of Iceland’s non-explosive volcanoes, Bardarbunga, underwent the largest basaltic eruption since Laki in 1783. The erupted lavas put chlorine and bromine into the atmosphere as hydrogen chloride and hydrogen bromide, which had the same effect on the ozone layer as CFCs from spray cans. They depleted the ozone layer and admitted more solar UV-B. Now that this event is over, things can, and probably will, return to “normal,” i.e., to “hiatus” conditions.
It should be noted here that explosive, andesitic volcanoes are well known to produce global cooling because they put a lot of sulfuric aerosols into the stratosphere, which reflect away sunlight. Together, explosive andesitic and non-explosive basaltic volcanoes have been causing global cooling and warming, respectively, throughout much of geologic history, as a careful review of the geologic record attests.
The bottom line is that the demonization of carbon dioxide is probably a mistake, and is unnecessary, as is the plan to replace all kinds of fossil fuels with less efficient, more expensive, environmentally destructive, non-renewable energy sources at great expense and socioeconomic disruption. The probability is that in going after CO2, we are going after the wrong target, and it will do no good. It seems far more important to make sure that the Montreal Protocol is assiduously followed, and that CFCs are closely monitored, going forward.
For the record, I am a dedicated progressive and no friend of any of the fossil fuel industries, in which I have no investments whatsoever. I am simply concerned with doing science right and (to borrow a metaphor from Cervantes) with saving us from falling flat on our face by tilting at the wrong windmill.
Note from Anthony: when I first received this essay, and saw the title, I thought perhaps it was another one of those “slayer” arguments. But after reading it, and looking into the background of the author, I’ve decided it has merit and is worth discussing. I did add one word to the title “significant” since the author notes in his essay that some warming is noted. From his personal page:
DAVID BENNETT LAING
Author, geologist, Earth systems scientist, ecologist, botanist, professor, sailor, alpine skier, retired, Aspie, Dartmouth ’62, Harvard ’72, progressive online activist, Penobscot Bay, Maine, Charlotte Harbor, Florida, Rio Bueno, Jamaica, park ranger, forest ranger, folksinger, copy editor, Seafarer 38 ketch, house/barn/cottage on 3.5 acres with ocean view, no debts, no regrets
Ozone loss causes warming? By that logic, Antarctica should be warming the most. It’s actually warming the least….
beng135: See Hughes, G. L., et al., 2007, Statistical analysis and time series models for minimum/maximum temperatures in the Antarctic Peninsula, in which warmings of up to 6.7 degrees C, the world’s highest, are reported.
Well, we’ve moved on from angstrom’s experiment…
Here’s a detailed review
http://www.realclimate.org/index.php/archives/2007/06/a-saturated-gassy-argument
http://www.realclimate.org/index.php/archives/2007/06/a-saturated-gassy-argument-part-ii/
I’m not much into atmosphere based climate science. I’m an ocean man.
A good, thoughtful article. And I think the sidetone on ozone/UV is interesting. The “pause” in global warming does parallel a stabilization in stratospheric temperatures following the 1996 full-implementation of the Montreal Protocol.
Some worthwhile points in this article to ponder.
Tx, Steven Capozzola!
The value of very long instrumental data series
by Alan Longhurst https://judithcurry.com/2016/10/10/the-value-of-very-long-instrumental-data-series/
AL shows the “big jump” in temperatures from 1985 to 1990. Up one step in the stairs.
The Trend Profile of Mean Global Total Column Ozone 1964-2009
by Jamal Munshi http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2843032
JM shows the great decline in ozone between 1970 to 1993. Down in the cellar.
Most of the decline in ozone comes earlier then the jump in temperature. And ozone depletion stopped in 2003 when temperatures stabilized. Why is climate science so reluctant to analyse the additional energy from UV radiation.
From: Peter Ward. http://ozonedepletiontheory.info/ozone-distribution.html
“When total column ozone is depleted, less solar energy warms the stratosphere causing the tropopause to rise, allowing more UV radiation to be absorbed by Earth. When total column ozone increases, more solar energy is absorbed in the stratosphere, warming the stratosphere, lowering the tropopause, cooling Earth. Ozone accumulates over the Arctic in winter (Fioletov, 2008), causing increased heating of the stratosphere and less heating of Earth resulting in minimum temperatures that are particularly cold during January in the Arctic and at mid-latitudes when Arctic air masses move southward. Depletion of this polar ozone peak warms minimum winter temperatures and it is this warming of minimum winter temperatures that is most clearly observed. The extremes of temperature are much larger during the winter when ozone accumulates and is depleted. For example, in Britain, the difference between record high and record low temperatures from 1870 through 1999 was 47oC during February-March and 38oC during July (Webb and Meaden, 2012).”
“The area of the Antarctic ozone hole grew rapidly from the 1970s to the 1990s while minimum ozone in the hole decreased (NASA Ozone Hole Watch).”
“On land, much of the energy absorbed during the day is radiated back into the atmosphere at night. At sea, however, ultraviolet radiation penetrates the ocean to depths of more than ten meters (Tedetti and Sempéré, 2006), making it more effective at heating the ocean than infrared radiation absorbed near the surface where much of the energy is lost back into the atmosphere at night. Similarly more ultraviolet energy is reflected back into the atmosphere on land than at sea. The effect of ultraviolet radiation on air temperatures could be small unless there is substantial lower-tropospheric pollution including O3 and NO2, while its effect on increasing ocean heat content could be substantial. The oceans account for 93% of the warming of the Earth system that has occurred since 1955 (Levitus et al., 2012).”
Just a point to consider: Essentially all the UV is completely scattered by the atmosphere (Rayleigh scattering). Half of it goes to space anyway. This means that most of the UV radiation that falls on the Earth is at a wide range of incidence angles, due to the scattering (same story for the blue light of the sky). Low-incidence light will tend to specularly reflect from water surfaces (glare). I don’t see much discussion of these facts, so I am doubtful of the conclusions drawn in their absence.
While varying amounts of UV get to the surface, and overall its share of TSI is small, its flux does vary far more than TSI. As what does make it through the ozone layer and other obstructions heats ocean surface, the big variation in the UV spectral component of TSI could well be IMO an important player in climatic fluctuations.
Chimp: especially since solar UV-B has 48 times the frequency of Earth’s IR absorbed by CO2, and is correspondingly more intense as a heating source.
This anomally is similar to some of what I’ve been doing. It peaks in March in the NH as that’s when the temp changes the most per day for warming, and oct/nov it’s changing the most negative per day. I could see Co2 rising as plants decay (you mentioned this) falling as plant life starts to grow for spring.
This rate of change has changed only slightly since the 50’s.
@ur momisugly micro6500
Your above mimicry of “junk science” propaganda doesn’t help your credibility any.
Here, let’s see iffen you can contemplate, correlate or associate the increases & decreases of the plotted “temperature and CO2 ppm” on this graph, to wit:
http://i1019.photobucket.com/albums/af315/SamC_40/1979-2013UAHsatelliteglobalaveragetemperatures.png
“Now, if variations in the concentration of CO2 in the atmosphere had any effect on temperature, you would expect that the peak in CO2 would occur with or before the peak in temperature, but it actually occurs two months afterward, which shows that variations in atmospheric CO2 cannot possibly have a significant effect on temperature.”
logic fail.
study the theory harder.
Steven Mosher: Theory must ALWAYS be constructed on the basis of hard data; not the other way round.
David, I really like your one-sentence CV that Anthony unearthed. It is original and says everything that needs to be said. Mine differs from yours in schools attended except for Dartmouth which I finished 17 years before you did. I want to talk here about the Keeling curve but first let me tell you that you made a huge mistake when you chose 1975 – 1998 for a comparison period. NOAA and associates have falsified this segment by creating a phony warming when the eighties and nineties were clearly a hiatus (no warming) period. Figure 15 in my book “What Warming?” shows this hiatus clearly. There are five El Nino peaks there that line up horizontally. NASA described this temperature segment in 1979 as follows:
“Unlike surface-based temperatures, global temperature measurements of the Earth’s lower atmosphere obtained from satellites reveal no definitive warming trend over the past two decades. The slight trend that is in the data actually appears to be downward. The largest fluctuations in the satellite data are not from any man-made activity, but from natural phenomena such as large volcanic eruptions from Mt. Pinatubo, and from El Nino. So the programs that model global warming in a computer say the temperature of the Earth’s lower atmosphere should be going up markedly, but actual measurements of the temperature of the lower atmosphere reveal no such pronounced activity.”
Note that actual measurements fail to show any warming and yet the masters of global temperature curve have inserted a fake warming in there. And it does not stop there but is continued on to the 1998 El Nino and to the twenty-first century that follows. I have been warming about this for years but these big shots at NOAA etc. feel that they are so important that they can ignore any pipsqueak objections from climate workers. And their supporters like Bob Tisdale have been libeling me about this but Anthony Watts does nothing. I knew that somebody working on temperature would eventually be misled by this falsification and here it is. To me, it is a scientific crime and they should be thrown out.
I am glad you took up the wiggles of the Keeling curve because they prove that no warming shown on global temperature curves can possibly be greenhouse warming. Comparing the amplitude of the wiggle in ppm with the amount of CO2 shown on the carbon dioxide curve in your graph we get ratios like 60 to 1 (for 360 ppm) etc. A more important comparison is between the Keeling curve and the Global temperature history curve that NOAA, GISS and others put out. They differ but not in the important aspect of comparison with Keeling. On the scale where global temperature ups and downs are shown clearly the seasonal wiggle cannot even be seen on the Keeling curve except as a fuzz that is barely noticeable. My first estimate in comparing the amplitudes of a substantial global warming with Keeling was that it is probably a close to a hundred to one ratio. We are told that the warming shown on the global warming history graph is caused by the greenhouse effect of carbon dioxide. Well, put it next to the Keeling curve that shows what carbon dioxide actually does and you immediately note that none of the temperature peaks and valleys on the temperature chart have any counterparts on the Keeling curve. It is totally smooth, except for that fuzz after 1958. I actually noticed it many times but for reasons unknown I did not give it any thought. What I missed was the fact that if a warming starts from scratch, atmospheric carbon dioxide displayed by the Keeling curve also must increase at the exact same time to make greenhouse warming possible. Since this does not happen it follows that the warming that we observe cannot possibly be greenhouse warming. Actually, there is a small amount of greenhouse warming in the forests of the world as shown by the existence of the seasonal wiggle. Its amplitude is a hundredths of the amplitude of the global warming peaks they try to sell us as products of greenhouse warming. If that is really the case these warm peaks must have been created by a carbon dioxide cloud a hundred times larger than we get from dropping of leaves that created the wiggle. And also invisible at the same time too because the Keeling curve cannot see it at all. In short, greenhouse warming is impossible and decarbonization of atmosphere that is forced upon us cannot possibly change the climate. With that, AGW is dead and I want a refund for all my taxes applied to this asinine effort to change the climate.
Typos:
17 years =11 years
warming about = warning about
Arno Arrak: Thanks for the input, Arno, but if you say temperature didn’t shoot up during 1975 to 1998, then why do all the temperature time series besides NOAA’s and the satellite records also show that it did just that? Cheers, David
If the anomaly in a given month/year is just the average temperature for that month/year minus the average temperature for that month across all years of the baseline period, would the curve have the same shape if a different baseline period had been selected? And if not, how can you convince anyone that this is the one true shape?
If each point on the graph represents an average over more than 2 decades, how does that point represent the long-term change over that period?
Anthony’s first instinct was correct – this is the kind of creative accounting employed by so-called “slayers”.
ScottM October 11, 2016 at 1:31 pm
The way you define it the shape of the curve would not be affected but the vertical placement of the entire curve would be. As an example, he just accidentally picked a baseline containing a fake warming and as a result his temperature curve stands higher on the graph than it would be if the unmodified baseline had been used.
The way I define it? The way base period is defined by scientists, in order for this curve to have anything other than a flat line at zero, the anomaly must be different during the period over which the graph is averaged than it was over the baseline. What if we selected 1975-1998 as the baseline, and then chose to graph the anomaly of each month over the period 1981-2010 (NOAA’s baseline). The effect will be to replace each X-Y with Y-X. The curve would be inverted from what you see above. Now pick a baseline of 1950-1980 and redo the graph for 1975-1998 relative to that baseline. The effect will be to replace each X-Y with X-Z. Unless all of the Zs are equal to the Ys (of the respective months) minus a constant, your conclusion that only the “vertical placement” would be affected is unfounded. You seem to think that they are *necessarily* equal. But if it were necessarily the case, then it would also necessarily be the case that all of the Xs would be equal to the respective Ys. And that would yield a perfectly flat graph, which we do not see above. That the author’s graph doesn’t have a flat average temperature anomaly curve therefore disproves his own case. The curve shows nothing. Note: For a given month, X = average temperature in that month from 1975-1998; Y = average temperature in that month from 1981-2010; Z = average temperature in that month from 1951-1980. Anomaly is the difference in a month’s average temperature for the period of interest and the same month’s average over the base period; e.g., X-Y or X-Z depending on which of two base periods we choose. In general, X, Y, and Z for a given month can (and most often will) have different values because of the different periods represented.
Sorry, I meant to write “the way anomaly is defined by scientists” in that opening paragraph.
I think NOAA’s baseline is the whole of the 20th century, so we would expect to see some difference if we only take the mean over a few decades within that period (of course scaling it from 0-100% means we don’t know the magnitude of the cycle). Basically this just shows how the seasonal cycle over that period was slightly different from the seasonal cycle on average over the 20th century.
More importantly, taking the baseline month by month in this way to calculate the anomaly will subtract off any influence the seasonal cycle in CO2 might have on the seasonal cycle of temperature before the analysis is even performed. It is hardly surprising that the analysis doesn’t find an association given that it analysed data with the seasonal cycle deleted already!
ScottM: Good points. No, actually, the curves would look the same if different base periods were used. The resultant anomalies are referred to certain base periods for convenience in discussion, only. Its sort of like drawing the contours on the head of the Statue of Liberty. You can start at the base and include everything from there, or you can take your base line at neck level instead. The contours would come out exactly the same either way.
As to your second point, I removed all long-term trends from the data sets by averaging monthly values, so my graph contains no long-term trend information whatsoever. It only assesses the question of the relative timing of maximum variabilities in each of the datasets.
“ScottM: Good points. No, actually, the curves would look the same if different base periods were used. ” no they wouldn’t becase the anomalies are relative to the mean for each month of the baseline period, not some constant temperature.
dikranmarsupial: I stand corrected on that. They would, in fact, look different, and the graph of temperature variation from a simple 20th century norm probably wouldn’t show the apparent dependence of temperature variability on ozone depletion as clearly as the deviation from a monthly norm does.
This is so easy to spin the opposite direction. I haven’t read every comment, so I don’t know if anyone else thought of this, but here goes…
If you look at the monthly anomaly curve, CO2 maxes in May, and temps in March. Rather than CO2 tailing temps by two months, it’s actually LEADING temps by ten months.
Q.E.D.
Where’s my grant money?
James Schrumpf: I did think of this possibility, but it seemed so remote that I didn’t consider it seriously. Can you give me one good reason why a signal in one dataset should take ten months to show up in another one?
I don’t have any idea, I was speaking totally tongue-in-cheek.
Obviously the next step is to look at the daily cycle. During the day temperature rises as CO2 levels fall. So CO2 must cool the planet. Obviously.
Note for the sarcasm deprived:
Obviously … this is a complete load of rubbish, just like the original article.
Figure 2 shows both temperature and ozone anomalies coinciding with the spring solstice when insolation is max, as is to be expected. Also, the decline in CO2 waits for May when the plants awake and begin their rush of growth, hungry for all the CO2 we humans released over the winter. But there is no need to invoke irrelevancies such as the CFC boondoggle. These made-up scares serve only to illustrate the perfidy, mendacity, and utter dishonesty of a few scientists with their many greedy hangers-on.
pochas94: How else would you explain the sharp coincidence of the warming and ozone depletion curves and their high correlation than that they suggest that ozone depletion is a serious possible contender for an important role in global warming? Are you objecting to my agreement with Molina’s mechanism for the photodissociation of CFCs on polar stratospheric clouds releasing monatomic chlorine, which then destroys ozone catalytically?
Right. May is an equinox, not a solstice.
Nope. Neither.
I wuz confused.
Neither the argument for carbon dioxide levels or ozone layer changes being the cause of global warming is supported by the fact that the global warming is not continuous but rather seems to occur in a series of rises and pauses that do not appear to be random but rather occur in a predictable pattern.
Looking at the last thirty years, it was below the trend line for the first ten, above the trend line for the second ten, below it for the last ten. And has recently returned to sit on top of the trend line, which is maintaining a positive slope.
William Everett:
Since 1870, there have been 2 depressions, and 28 recessions. ALL are coincident with temporary increases in average global temperatures, due to fewer sulfur dioxide emissions into the atmosphere because of the reduced industrial activity.
Between 1972-present, Clean Air efforts have resulted in decreased SO2 emissions, which also cause increases in average global temperatures.
There is no cyclic nature to climate change. It is all due to the amount of SO2 aerosols in the atmosphere.
Bottom line: Climate change is being caused by the EPA
Burl Henry, I am not talking about climate change. I’m talking about temperature history and it does occur in a 500 year cycle according to the currently held view of notable climatologists. I’m saying that there appears to be a thirty year cycle of warming and pause in warming. As far as ScottM’s comment, going back thirty years from the present is meaningless. The changeover point between the previous warming period and the current pause occurred around 2002 by my reading of the temperature record. If there is a pattern to the temperature change then the next change point from a pause to a period of warming should occur around 2032. The global mean temperature at that time should be about the same as it was in 2002 barring the occurrence of an El Nino effect at the 2032 changeover time. Also, if the apparent pattern holds then there will only be 40 years of warming in the present century.
William Everett:
You mentioned “The changeover point between the previous warming period and the current pause..”
A couple of questions:
What years are encompassed by “the previous warming period?
Temperatures in 2014-2016 are substantially higher they were in 2013 and earlier. What do you mean by “the current pause”?
William Everett, you missed my point in a number of important ways, by focusing simply on the apparent cyclic aspect of what I wrote. The same inference can be drawn whether the wander above and below the trend line is cyclic or simply a random walk: namely, that regression to the trend line happens. But you also included a non sequitur that temperatures would be the same in 2032 as they were in 2012 (if we stipulate that the wander is truly cyclic). You completely ignored the fact that the wander is superimposed on a positive trend – so when you complete a cycle, you will be at a higher temperature than you were at the start of the cycle.
If you read carefully, you will not find that I hypothesized a cycle of any kind; merely described what the temperatures did.
William Everett: One important consideration is that stratospheric ozone concentration is quite evanescent, fluctuating rapidly during ozone’s short lifespan of several days, whereas the atmospheric concentration of CO2 is quite steady. This is consistent with the temperature effects being erratic.
The years in the previous warming period were from about 1974 until about 2002. Temperatures in 2014-2016 were caused by a strong El Nino effect which has subsided. There have been other temperature spikes since 1880 caused by El Nino effects. These short term effects have not erased the longer range of temperature change which features periods of warming and periods of pause in the warming. Refer to the global temperature record published by NOAA and GISS.
William Everett:
For the period 1972 to 2000, global Sulfur Dioxide emissions fell by 25 Megatonnes, which resulted in O.5 deg. C. of anomalous temperature rise in land-ocean surface temperatures..
Between 2000 and 2013, global Sulfur Dioxide emissions fell by 8 Megatonnes, for an anomalous temp. rise of 0.16 deg.C.(the “pause”, although they never did actually stop rising).
Your cycles have to occur for a reason, and it appears that they are .related to the amount of SO2 emissions in the atmosphere. Between about 1955 and 1972 SO2 emissions were rising, which would cause cooling, and your previous “down” cycle probably spans those years.
Going forward, I believe there should be an “up” cycle, and never a down cycle beyond some La Nina cooling, unless we have some large volcanic eruptions, because of continuing reductions in SO2 emissions.
Burl Henry: As I noted previously. Peter ward found that there was a twenty-year lag between the anthropogenic rise of SO2 in the atmosphere and the rise in global temperature, but the rise in ozone depletion curve only lagged temperature by some five years. He therefore concluded that ozone depletion was a more likely cause than SO2.
William Everett: Peter Ward attributes the warming spike in 2015 to 2016 to ozone depletion caused by halogens emitted during the eruption of Iceland’s non-explosive, basaltic volcano Bardarbunga from August, 2014 to February, 2015, the largest such eruption since Laki of 1783.
I believe that the reason for the 500 year warming (or cooling) cycle is because of a change in Earth’s relationship to the Sun. I think the apparent 30 year pattern of warming followed by a pause in warming is also caused by change in the Earth’s relationship to the Sun. The sulfur dioxide emissions are probably in the same category as carbon dioxide emissions, insignificant as an influence on global temperature.
William Everett:
AT this time, I cannot speak to earlier cycles, but over the past 40 years temperature projections based solely upon the amount of reductions in SO2 emissions are accurate to within .02 deg. C., or less.
For this time period, they are highly significant with respect to average global temperatures.
Burl Henry: Could be a spurious correlation.
Until it is shown otherwise I think that the mean global temperature will remain level until about 2032 reflecting the temperature history available to us since 1880.
William Everett: On what data do you base your assumption that temperature will remain level until 2032, and what happens then?
Schrumpf and Ferdinand make some valid points above, which I’ll not comment further about.
The main article, really doesn’t show what it thinks it does. Some things to look at:
* You’ve ignored the fact that both temperature and CO2 respond to a third variable (Sun)
* Does the answer change if the period is changed? 1998 was quite a remarkable outlier. If the start/end points are moved by 10 years, do your answers change?
* 24 years is a very peculiar time span to look at for a climate time series. What happens to your answer if the period is made 20 years, or 30 years instead?
* What point of science dictates your choice of time span rather than to use the entire period for which monthly data are available?
* ‘normalized’? Why not use ppm and K for your analysis? They’re much more physical. The ‘normalization’ process itself could be producing the timing of the peaks.
*From your description, you’re using only those 12 data points for your analysis of T vs. CO2. And have not looked at all at the variances in the monthly estimates.
* More informative will be to look at the monthly anomalies of T and CO2 (un-normalized) through the period of record.
A point mentioned by Ferdinand is important. Time scale matters. Nobody expects the climate system to respond instantaneously to changes in CO2, or anything else. Even if it did, you couldn’t tell instantly because there’s so much natural variability.
If you examine glacial ice at short time scales, it is absolutely a solid. You can even send acoustic shear waves through it, one of the hallmarks of a solid. But that’s looking at short time scales (seconds to hours). Over long time scales (years to decades and up), it is a fluid. You see that in the fact that ice flows downhill, flows out in to ice shelves, and the front of the ice shelf advances through time.
The mathematical analysis which would let you address this issue is spectral coherence, where you’ll find the phase (lead/lag) relationship, as well as the correlation, in terms of the time scales. Where there is no coherence, phase relationships are meaningless. Very likely this is where your data are. But at decadal scales you might see something else.
Co2 changes to temperatures. 1998 wasn’t unusual.. plotting the temperature and co2 anomolies per year instead of the total in the air and temperature gives a very clear picture that over the last 60 years while there has been a slight warming trend, there is no reason for co2 to track with downward movements in temperature as co2 production increased. The warming trend is evident in the graph where the total co2 is given and temperature is given in year to year tenths of a degree. Co2 anomolies and temperature track every year (except for changes in the last 5 via NOAA ) which is a different graph that isn’t published. 1999 wasn’t a variation. Neither were the years that followed. Every year since 1998 the co2 levels did not reach the same peak even though during that time period production of anthropogenic co2 increased a billion metric tons year each additional year. Only this year with an el nino did the co2 levels exceed 1998, and from what I can see, it was temperature. Co2 played no role in controlling temperature.
Further, co2 anomolies track cosmic ray and solar activity reaching a peak then subsiding during the solar cycle. It’s like clock work. The only thing the AGW crowd has proven is that it’s slightly warmer and not due to co2, and certainly not dramatically .
What’s making it warmer ? I am uncertain. However, I can rule co2 out. And the scientific community will too. Whether it it’s colder or not won’t be the determining factor.
In fact Ferdinand brought that up that it was a variation from 1998 to 1999, so I did all the years. Co2 varies with temperature. It’d be a lot warmer if it were the other way around. That I am also certain of. We wouldn’t be having this discussion.
Also if the analysis uses the northern hemisphere temperature anomalies, any effect the seasonal variation in CO2 on temperature will already have been deleted from the temperature data as the anomaly for January is the anomaly relative to the January mean, the anomaly for February is the anomaly relative to the February mean, etc. So the analysis is incorrect from the outset, even without all of the other concerns.
Indeed, if you plot the real CO2 (and δ13C) changes over the seasons, the change at ground level is quite high and more smoothed at the 3400 m heigth of Mauna Loa, here plotted over two periods, :
http://www.ferdinand-engelbeen.be/klimaat/klim_img/seasonal_CO2_MLO_BRW.jpg
It looks like that the more recent biological carbon cycle increased somewhat, but that is barely visible at Mauna Loa. Moreover, the peak difference at Barrow is in August, not in May, where the anomaly should get around -2 ppmv in that month… I wonder what the anomaly percentages in Figure 2 in real life figures are…
dikranmarsupial: On the NOAA website https://www.ncdc.noaa.gov/cag/time-series/global, at the top, it clearly states that the anomalies are “with respect to the 20th century average.” Below, the user is given various choices of how to display the monthly record. As a matter of fact, though, my graph would have given substantially the same information as it did anyway if the temperature anomalies had, indeed, been deviations from monthly averages over a certain base interval.
dikranmarsupial: Sorry. I just clarified with NASA that you are quite right, that any monthly anomaly is a departure relative to the normal value for that particular month over the base period. This is actually better than I thought it was for my purposes, because it shows the maximum and minimum variabilities for temperature much more clearly than it would have if the average to which it referred were a simple mean over a base period. Thanks!
David, no it couldn’t be worse for your analysis because most of the effect of the seasonal variation in CO2 would be in the monthly means that are subtracted off to compute the anomalies. The anomaly doesn’t tell you anything at all about the effect of seasonal variations of CO2 on seasonal variations on temperature. What you see in the graph is the difference in the average seasonal temperature over the period of interest and the average seasonal temperature variation over the baseline period. There is no reason to think that is caused by some change in the seasonal variability in CO2.
dikranmarsupial: I think I see, now, what you’re driving at. Because the temperature anomalies during the analysis period are departures from averaged monthly values during the 20th century base period, you’re saying that any seasonal effect of CO2, or any other influential factor, on temperature variation would have been removed by the averaging process, and therefore it would not show up in my graph. I think that if that were, in fact the case, then the temperature anomaly curve in the graph would not have shown such clear seasonality, peaking strongly in March, which is counterintuitive. Why March, and not July, for example? The fact that ozone depletion also peaks in March, and is so well correlated with temperature anomaly, suggests that it, rather than variation in CO2 concentration or high Sun angle, is likely the main driver of temperature anomaly. Nonetheless, I will check the average monthly record of temperature (not anomalies) over the 20th century to see whether the peak of CO2 concentration in May has any kind of seasonal effect. If so, I would expect it to take the form of an early broadening of the seasonal peak from its expected maximum in July back toward May, and I will post the result in this forum. In any case, I want to thank you sincerely for being my smartest and toughest critic on this, and for forcing me to think carefully about my analysis.
dikranmarsupial: I did a graph of NOAA’s baseline average monthly global temperature values for the 20th century used to compute monthly temperature values, and it turned out to be a simple sinusoidal curve with a symmetrical peak in July (several attempts to upload the graphic failed, but the monthly values are 12.0, 12.1, 12.7, 13.7, 14.8, 15.5, 15.8, 15.6, 15.0, 14.0, 12.9, 12.2). No effects other than the obvious one of Sun angle are evident in the graph, which suggests that 1) there appears to be no significant long term effect by any forcing agent on baseline monthly temperature in the northern hemisphere, and 2) my 1975 to 1998 graph succeeded in selecting short term variabilities in the three variables, and therefore gives a realistic picture of the effects of CO2 and ozone depletion on monthly temperature anomalies.
I suspect normalization of the temperature anomaly difference was done to hide the fact that the 0 to 100% range actually represents a very small difference range. There is enough overlap between the base period and the chosen observation period that I would be surprised if this amounted to more than a swing on the order of 0.1 K. More evidence of slayer-style creative accounting.
The normalization itself varies seasonally, it turns out. So these curves themselves say almost nothing about correlation and phase between T and CO2. They say at least as much about what times of year have the least variation in those variables. And science has long known that T depends on more than just CO2, and CO2 depends on more than just T.
ScottM: No, I normalized simply in order to render the three curves at the same scale. The important thing is the shapes of the curves, where they peak and trough and what they do in between (deflections, etc.). Sorry, but I am not a “slayer'” (whatever that may be). I’m simply asking questions.
Robert Grumbine: In answer to your points,
1) Right. I was only asking the question “Can variations in atmospheric CO2 concentration cause variations in temperature?” and the second question “Can ozone depletion be considered a viable candidate for a cause of global warming?” The answers to these two questions were no and yes, respectively.
2) through 4) The interval 1975 to 1998 that I chose was the time when 1) temperature shot up at a fast rate (i.e., something important was happening then) and 2) when anthropogenic CFCs were massively introduced to the atmosphere. The evidence might have been rather ambiguous if I had chosen a different analysis period, and my conclusions would have been more tentative.
4) I normalized (put the highest value of each data set at 100% and the lowest at 0%) so that I could compare the three curves on the same graph rather than complicate matters by providing a different scale for each variable. This transform did not in any way affect the shapes of the curves.
5), 6), and 7) Correct. I averaged the monthly values in order to remove trends and other variations so that I could arrive at a general conclusion of how, on average, the curves behaved, month-by-month, during the time interval chosen.
In Nature, including the atmosphere, responses to stimuli are usually immediate. Otherwise, a memory storage component is introduced, which only complicates the system in question.
When rheidity (as that of ice) is introduced, naturally response times are lengthened, but the atmosphere has very low rheidity, so response times to stimuli are usually immediate.
Mathematical manipulations of compiled data are intriguing and sometimes even compelling, but they can very often lead to false conclusions, as I have found over my 55 plus years of doing science. For this reason, I tend to use mathematics and theory as tools to help explain the configuration of real data when there is doubt as to what they really mean. A case in point is my taking monthly averages of the three data sets to remove trends and nonseasonal variabilities in this analysis in order to focus on two particular questions of interest.
1) You selected a cherry-picked time, and did nothing to see whether your cherry-pick affected your answers. Further, it’s a time you consider to be unusual (that rapid spike) in temperature, which immediately makes it suspect as to being representative of climate.
Your transform of the data affects your answer. Whether it’s standard in other fields for other reasons is irrelevant. Having performed multiple transformations, it’s on you to demonstrate, not merely assert, that the transforms do not affect your results.
Since your transform is itself time-varying, it certainly affects your results.
It’s ironic that you talk so negatively about ‘theory’, this analysis is laced with theory — not least the theory that your analysis is representing the climate system. Further, the problematic issue Schrumpf raised and which you don’t see. It is only by theory that you can select between 2 month lead vs. 10 month lag. While claiming no theory on your part, you apply one here.
Don’t see what relevance your claim of 55 years in science has to anything. Not least because you’re quite hard to find in any of the abstract databases. Doing science for a half century normally leaves more tracks than that.
In any case, I’m minded strongly of Feynman’s comment about doing science. The easiest person to fool is yourself, so you have an obligation to bend over backwards (his term) to avoid fooling yourself. In this case, you should be doing work like checking how your answers change with different (and longer) time periods, without your multiple transforms, and so forth.
On the other hand, since I see you reject photons, which were also a central part of his professional work, I suppose there’s no surprise you reject his principle here, too.
Robert Grumbine: I’ll gladly answer your objections because responding to sharp criticism helps me to refine my thinking on the matter.
First, yes, I did “cherry pick” the analysis period for two reasons: first because it represented the most dramatic increase in temperature in recent time, and therefore I concluded that “something important was happening then,” so that fact would make 1975 to 1998 an important time to look at. Second, I wanted to test Peter Ward’s contention that anthropogenic chlorine was an important player in global warming through its thinning effect on the ozone layer. Thus, my cherry-picking did, indeed, affect my answers, and I hope that you’ll agree that because of these two reasons, it was to good effect. As for the period being “representative of climate,” I regard that as irrelevant. I specifically eliminated long-term effects from the analysis by rendering the temperature anomalies as monthly averages. I was only interested in the question, “Can month-to-month variations in atmospheric CO2 concentration have a significant effect on temperature in the northern hemisphere?” By showing that CO2 variation peaked in May, whereas temperature anomalies peaked in March, two months earlier, I think I answered that question in the negative, except for a very minor possible effect in June.
No, my objectives in transforming the raw data were to focus on the information therein in the most efficient manner, specifically 1) to remove long term effects from the analysis so as to concentrate on the causality question above, and 2) to render the three curves represented at the same scale on a single comparison graph as percentages instead of providing a separate scale for each curve, which I felt would be unnecessarily confusing (perhaps I was wrong in assuming this!) These transformations did nothing whatever to change the information contained in the raw data. I hope that you find this is an adequate defense of my methodology.
No, the time-varying aspect of my analysis doesn’t affect my results at all; in fact it reveals them in that it clearly shows that CO2 could not influence temperature because its peak concentration occurs two months before the peak concentration in the temperature anomaly.
Again, no. What I object to in theory is when people come to some general conclusion about reality when they lack the solid data from the real world to back it up. Such is the case with Arrhenius’s 1896 geometric/arithmetic argument concerning CO2 and temperature. In 1900, Angstrom tested it for him and came up with essentially negative results. The theory has never been tested against hard data since except for my northern hemisphere analysis here, which evidently shows, on the basis of real data from the Earth system, that Angstrom was right, after all.
Frankly, I’m surprised that you succeeded in finding me at all in abstract databases. I am a polymath and a synthetic scientist (as opposed to an analytic one, although this is, indeed, an analysis), with a broad range of interests, and the normal peer review publishing process detests polymaths. I realized this about myself fairly early in my career (I also have Asperger’s syndrome, which actually explains a lot), and so I made a conscious decision not to finish my PhD work and to be content instead with my masters degree from Harvard rather than fight the peer-review system that likes narrowly-focused studies that are in line with accepted dogma, and I’ve not regretted the decision. Therefore, instead of finding me by journal articles, you’re more likely to find me under books, because independent publishing gives me a good opportunity to “go where no man should presume to go.” On the one hand, this should raise a big red flag in that no one but me provides a check on my ideas, but that’s not quite true. As an Earth systems scientist, I work from two principles: 1) that I should always base my conclusions on an Occam’s razor interpretation of real data from the Earth system, and 2) that connecting the dots is a powerful tool. In other words, if any of my conclusions is consistent with a broad range of real Earth data, then it’s probably correct. I then present my ideas on fora, such as this, or send them to key people in the discipline for their comments, or both. I pay close attention to feedback I get from both sources, and that keeps me honest (I hope!)
You’ll be thrilled to know that I’m taking Feynman’s good advice and subjecting everything I do to the hairy eyeball. I always test everything I come out with with the question, “Does this make sense in terms of all that I have already learned from Earth and from other people’s data-based research on Earth?” If yes, I proceed to seek out other hairy eyeballs, such as yours, to ferret out possible flaws in my methodology and conclusions, so thanks!
David Bennett Laing: “First, yes, I did “cherry pick” the analysis period for two reasons: first because it represented the most dramatic increase in temperature in recent time, and therefore I concluded that “something important was happening then,” so that fact would make 1975 to 1998 an important time to look at.”
You realize you have just described the Texas Sharpshooter Fallacy, don’t you?
I cannot understand why March would be the month with highest northern hemisphere temperature anomaly. Seems like it should be July or August
The analysis uses the northern hemisphere anomaly temperature dataset, which already has the normal seasonal cycle subtracted from it. Basically you pick a baseline period (which I think may be the entire 20th century) and compute the mean temperature for each month of the year. You then subtract those monthly means from the corresponding months in the dataset to give you the anomaly. However if you pick a period that is not exactly the same as the baseline period, then you will get a non-zero signal. The “anomaly” that is plotted just shows how the mean seasonal variation over period of interest differed slightly from the mean seasonal variation over the whole baseline period, and has nothing whatsoever to do with the seasonal variation of CO2. This of course immediately invalidates the the analysis.
thanks for this answer. I agree with you the analysis seems wrong if anomaly is calculated this way, although I still believe evidence is strong for very low sensitivity with zero or negative feedback
I disagree, paleoclimate (e.g. PETM) for a start is very difficult to explain if climate sensitivity is very low, I don’t see any strong reason not to accept the IPCC range, but the key point is that this analysis is fundamentally flawed, due to the use of the anomalies, as well as for the other reasons outlined above by myself and others.
Dikran,
The PETM is probably caused by a huge meteor impact, which increased tempertatures and CH4/CO2 levels due to burning forests and increased seawater temperatures, direct release of clathrates at the impact place and increased volcanic activity. One can’t tell from that impact that CO2 was the main driver for the temperature changes… See:
https://wattsupwiththat.com/2016/10/13/study-extraterrestrial-impact-preceded-ancient-global-warming-event/
Have a look at the end of the previous interglacial (the Eemian): CO2 remained high while temperatures (and CH4 levels) dropped to a new minimum and ice sheets did grow to a new maximum. Only then CO2 started to drop some 40 ppmv. That had no clear influence on temperature or ice sheet growth. The uncertain relative timing doesn’t play a role in this case as CH4, CO2 and N2O (a proxy for global ice sheet volume) are measrued in the gas phase and CH4 directly follows the δ18O changes (a proxy for SH ocean temperature) in the ice. Here for the Vostok ice sheet:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/eemian.gif
Which shows that the impact of CO2 on temperature and ice sheet formation is minimal…
Ferdinand, I think it is a bit of a leap to say that the “PETM is probably caused by a huge meteor impact” given the abstract of the paper says “…indicating that an extraterrestrial impact occurred during the carbon isotope excursion at the P-E boundary.” (emphasis mine), i.e. the paper doesn’t say the impact was the cause of the CO2 excursion, indeed the actual paper doesn’t seem to say anything much about the cause of the CO2 excursion it at all. The CO2 excursion apparently took place over millenia, rather than suddenly, which also argues against a meteor having a big impact (if you will excuse the pun ;o). As far as I can see there is very little in the earlier WUWT article to support the idea that the rise in temperature during the PETM was cause by anything other than an excursion of GHGs, which is hard to explain if climate sensitivity is very low.
I’ll look into the Emian episode you mention, however CO2 is not the only thing that affects climate, so we can’t draw strong conclusions from a single paleoclimate event in isolation unless we know that “all things were otherwise equal” on that occasion, we want the best explanatory theory for all that we see in paleoclimate, and at the moment AFAICS CO2 has a substantial role to play (mostly as a positive feedback).
Dikran,
It seems that most of the change was from seabed methane, probably from a huge meteor impact in an ocean.
http://www.pnas.org/content/113/28/7739.abstract
At current low methane emissions, the half life time to convert it into CO2 is only 10 years, but that depends of the amount of OH radicals in the upper troposhere, which are limited in production (by UV). Thus huge quantities of CH4 released at once need much more time to get destroyed.
CH4 is a much more potent GHG than CO2 and may be the main cause of the temperature increase.
Some beg to differ about the speed of the change:
http://news.rutgers.edu/research-news/new-finding-shows-climate-change-can-happen-geological-instant/20131003#.WAEj74Kt9Ey
That also looks like a huge meteor impact somewhere in one of the oceans (and therefore no impact crater visible on land)…
Dikran,
In addition, the article itself at:
http://www.pnas.org/content/110/40/15908.full?sid=58b79a3f-8a05-485b-8051-481809c87076
Shows the estimated quantities of low-13C released:
Wow! That is about eight times all what humans have burnt in the past 166 years, released in less than 10 years…
Remains to be found what the meteor impact itself was on earth’s temperatures, it seems quite impossible to me that temperatures did go up in less than a decade even not from a doubling of CO2 (or even extreme quantities of methane) in such a short time span, as the oceans’ inertia is much too large. The opposite may be more appropriate: the impact may have heated the oceans instantly high enough to give the CH4 and CO2 releases…
Ferdinand Engelbeen
October 14, 2016 at 11:36 am
Has the surface of any large ocean area ever actually been acidic?
Ferdinand, that appears to be the very paper that was criticised in the comment paper for which I provided a link, showing it couldn’t have taken place over a decade or so, but took millenia. I’m no expert on this, but the existence of Zeebe et al suggests we should be extremely cautious about attributing PETM to a metor strike (as indeed the authors are in the new paper, as I pointed out).
Chimp,
Not that I know and normally never, even if you burn lots of coal and oil and gas…
But 3000 GtC as extra CO2 in less than a decade, that is something different.
3000 GtC instantly released means an instantly 3.75 fold increase of the current 800 GtC in the atmosphere or 400 ppmv increased to 1500 ppmv.
I am not sure about what pH drop that gives in the surface, as most pH calculations of that period (-0.3 pH unit) still assume a long period (~10 kyear) of release, which makes it possible to have a total mix with the whole ocean depth, while a sudden release only has an immediate impact on the surface layer (average 100 m depth). I think that the surface layer may have been slightly acidic for a short (a few decades) period.
Dikran,
Zeebe follows the “models show” lead, that is based on no-impact, GHGs do it all. But now is proven that there was a huge impact, which may have triggered not only a clathrate release (and partly immediate burning), + forest fires (with reduced δ13C compared to methane) plus a heating up of the oceans if the impact was large enough…
On the other hand assuming that the sediment layers were yearly, based on oxygen isotope changes is a little triggy, as there may be other reccurent periods which can trigger similar changes: the solar cycle which changes rain patterns / river discharge in mid-latitudes, the ~60 year ocean cycles (PDO, NAO), the ~1000-1200 climate cycle (Roman period – MWP – current warm period)…
Without detailed confirmation of the age structure it remains controversial…
Will certainly be followed…
dikranmarsupial
October 14, 2016 at 7:26 am
Observed reality, ie ECS of ~1.0 to 2.0 degrees C per doubling of CO2 concentraion, overlaps with the unscientifically derived IPCC range of 1.5 to 4.5 C, but just barely.
“which may have triggered not only a clathrate release…”
I am happy to agree with may have triggered a clathrate release, just not that it probably caused the PETM. For me the meteor is one of many plausible hypotheses about the cause of the PETM, and I don’t want to jump to conclusions.
Regarding models, they are an embodiment of our best knowledge of the physics. If you can make a physical model that explains the observations, your hypothesis is better than a hypothesis for which you can’t construct a physical model, as it shows that your hypothesis fits in with our existing physical knowledge and so this shouldn’t be ignored. The other thing you can do with a physical model is to move on from correllation by testing whether the proposed mechanism can explain the magnitude of the effect.
My intuition on the direct heating is that much of the energy of the impact would be quickly radiated away (firstly by vapourising the water in the immediate area of the impact) and the thermal conductivity of water isn’t high enough to spread the heat through the oceans, rather than to boil it locally. It is an idea worth considering, but I would want to see the physics demonstrated first.
Interesting collection of papers though,including the authors response to Zeebe et al.
Dikran,
Indeed the response of the authors of the rapid release is worth reading too, as they show the flaws in the model(s) Zeebe e.a. used:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3970519/
Will certainly be followed by more discussion and hopefully more constraints by new discoveries…
thinking about this paper some more, it would be interesting to see same graph during the pause We should see lower zone depletion and lower temperature peaks, still in phase if Mr. Laing’s thesis is true. Am I right?
dikranmarsupial: I don’t think it invalidates my analysis at all. As you yourself pointed out, these anomalies are in reference to the normal monthly values over the base period, which is the 20th century. My analysis therefore shows a curve of relative sensitivity of temperature, which has its peak in March and a small uptick in June. The maximum concentration of CO2 over the analysis period comes in May, and is therefore two months later than the month (March) in which temperature is experiencing its maximum deviation from the average for that month over the base period. This implies that CO2 can’t be causing this maximum departure, but that ozone depletion could. This makes a lot better sense that if the anomalies were deviations of T from an annual mean.
Ferdinand Engelbeen: Notice that In the graph you present here, CO2 goes up following temperature, which one would expect from the exsolution of CO2 from a warmed ocean, but that it goes down again as temperature again plunges, this time consistent with solution of CO2 by a cooler ocean. Otherwise, CO2 has very little correlation with temperature.
dikranmarsupial (Oct 13, 9:37): What’s wrong with the interpretation that the PETM temperature increase was caused not by carbon dioxide but by ozone depletion by chlorine and bromine emitted during the massive eruption of non-explosive, rift-related basaltic lava due to plate tectonic activity between Greenland and Scandinavia at the time? This would require subaerial eruption, of course, but that would be a possibility, considering the present subaerial exposure of Iceland.
Nelson Smith: The graph shows that March is the month during which the greatest effect on global temperature is felt (i.e., the greatest variation from March norms), and that is a very good reason to consider ozone depletion as the cause of that maximum variation.
Ferdinand Engelbeen
October 14, 2016 at 1:44 pm
Thanks.
Benthic forams suffered a major extinction at the PETM, but surface dwelling shelly plankton, dunno.
As for coccoliths, “calcification and net primary production in the coccolithophore species Emiliania huxleyi are significantly increased by high CO2 partial pressures”:
http://science.sciencemag.org/content/320/5874/336
Ocean acidification in response to rising atmospheric CO2 partial pressures is widely expected to reduce calcification by marine organisms. From the mid-Mesozoic, coccolithophores have been major calcium carbonate producers in the world’s oceans, today accounting for about a third of the total marine CaCO3 production. Here, we present laboratory evidence that calcification and net primary production in the coccolithophore species Emiliania huxleyi are significantly increased by high CO2 partial pressures. Field evidence from the deep ocean is consistent with these laboratory conclusions, indicating that over the past 220 years there has been a 40% increase in average coccolith mass. Our findings show that coccolithophores are already responding and will probably continue to respond to rising atmospheric CO2 partial pressures, which has important implications for biogeochemical modeling of future oceans and climate.
Also wonder about the combo of sudden CH4 and CO2 increases on ocean pH.
Chimp,
CH4 has no influence on pH, but is converted into CO2 in the higher troposphere and in part by bacteria in the oceans… With a huge impact, lots of clathrates may have been burned immediately…
Thanks.
Some think that if sudden GHG rise accounts for the PETM, methane is a likelier culprit than carbon dioxide.
Chimp,
Ah the Ehux, responsible for the nice white cliffs of Dover and most of the South England underground and similar in French Normandy…
They thrived in much higher CO2 levels during the Cretaceous…
Nice pages of Ehux at: http://www.noc.soton.ac.uk/soes/staff/tt/eh/
Chimp: I find it strange that no one has implicated the NOx and SO2 that emerges from smokestacks and tailpipes along with CO2 in ocean acidification. CO2 is buffered, meaning that its acidifying effect is minimal, whereas these other oxides are not buffered, which gives them correspondingly greater acidifying power. Also, I understand that ocean acidification is more of a problem near continents than it is in the open ocean, which makes the land-based, anthropogenic nitrogen and sulfur oxides doubly suspect, yet it is never mentioned. Why is this?
They’re not mentioned because they don’t have the effect you suggest. Sulphate ions are a conserved species in seawater, meaning that it is at a constant concentration in the ocean. Nitrate is at such a low concentration that it would have no effect on ocean pH.
Phil.: How do you know this? Yes, sulfate is a conserved species, but this is sulfur dioxide, and the conservation of sulfate in seawater of itself doesn’t mean that the local addition of more sulfur dioxide won’t have a short-term effect, which will, in time, be reduced by the sulfate conservation effect. The same goes for nitrate. I see no sense at all in the statement that “Nitrate is at such a low concentration that it would have no effect on ocean pH.” If anything, I would expect that its low concentration in seawater would mean that added NOx would have even more of an effect.
David Bennett Laing October 19, 2016 at 4:32 am
Phil.: How do you know this? Yes, sulfate is a conserved species, but this is sulfur dioxide, and the conservation of sulfate in seawater of itself doesn’t mean that the local addition of more sulfur dioxide won’t have a short-term effect, which will, in time, be reduced by the sulfate conservation effect.
SO2 (or more likely SO3 by the time it reaches the ocean) will react with water to form sulphate ions instantly and the solubility product of species like gypsum would precipitate out. When seawater infiltrates freshwater gypsum precipitates.
The same goes for nitrate. I see no sense at all in the statement that “Nitrate is at such a low concentration that it would have no effect on ocean pH.” If anything, I would expect that its low concentration in seawater would mean that added NOx would have even more of an effect.
I take it that you have no understanding of chemistry? The principle process which controls the pH of seawater is the equilibrium between bicarbonate and carbonate, nitrate is less than 50 micromole/kg compared with ~2000 micromole/kg of bicarbonate.
Phil.: I take it that you have a bit of trouble with civil discourse. I have always assumed that such ad-hominem remarks are a certain sign of someone who is defensive and unsure of his facts. Both of your arguments are theoretical and based on equilibrium conditions. My question is, could there be a local, kinetic effect of these oxides on marine life before such equilibrium conditions as you mention could be established.
The chemical reaction time is order microseconds even for the (buffered) carbonate system, too fast for your stated concerns to apply.
That CO2 is buffered in the ocean does not mean that its acidification effects are minimal. Merely that you need more CO2 to get a given amount of acidification. The ratio is a lab result. Look it up before dismissing it.
CO2 is also emitted in far greater tonnage than SO2 or NOx. Find that ratio and compare with the chemistry, rather than dismiss these items that disagree with your wishes as ‘theory’.
see also Feynman regarding your obligation to not fool yourself. You’re spending zero effort to avoid that.
Robert Grumbine: Au contraire, I take Feynman’s admonition very seriously. In the interest of that, can you please identify for me the specific references that you refer to on this? Tx.
David: If you took Feynman’s admonition seriously, you would (for instance) have looked up the chemistry of oceanic CO2 buffering before dismissing CO2 as a source of ocean acidification.
Ref: Surely You’re Joking: Adventures of a Curious Character.
Robert Grumbine: I’m quite familiar with the ocean carbonate buffer system and how it works, thank you (BTW, why are you so quick to use ad-hominem remarks?; they’re usually a sign of insecurity in the person who asks them). I was simply asking the question as to whether SOx and NOx might also have some effect, and pointing out that the question hasn’t, to my knowledge, been addressed in the literature.
It isn’t ad hominem to observe that you don’t understand the chemistry. That’s just reality — you could not have had the questions you did if you understood the chemistry, nor could you have made the assertions you did.
Nor is it ad hom to observe that you aren’t following Feynman’s dictum given that you conflate you not being aware of what’s in the literature with what’s true or not. Your not being aware merely means that you have work to do to obey the dictum.
Rainwater condensing from cloud formations dissolves CO2 to form weak carbonic acid, it always has and it always will. It has been responsible for stalactites and stalagmites in caves over thousands of years. Rainwater of course is initially pure water in which there are few ionic species; seawater however is quite different and contains many soluble cations and anions. Sodium and magnesium, which will form stable crystalline solid bicarbonates, are present in abundance. Carbon dioxide in seawater yields salts such as sodium carbonate which are soluble in water and are hydrolysed in solution thus:
Na2CO3 + H2O = NaHCO3 + NaOH
and their solutions are in fact alkaline. For the anti carbon green warmist lobby and the BBC to pronounce that the simple addition of carbonic acid or dissolution of CO2 in seawater will make it acidic is nonsense and he clearly does not understand the complex ionic system pertaining in the oceans. The capacity of seawater to buffer pH changes is well known and its pH always remains in range 7.5 to 8.4 which is alkaline.
+1 on that.
David Bennett Laing October 12, 2016 at 4:01 pm
David, thanks kindly for your reply.
As to the short- and long-term relationship of CO2 and temperature, both of them have been discussed for years. The short-term effect is the one you discuss above—annual termperature-driven variations in the biosphere cause following and corresponding changes in CO2 level. These annual changes in atmospheric CO2 concentrations are on the order of a percent of the CO2 level or so. That effect we have lots of evidence for.
On the other hand, there is the long-term effect of CO2. This involves a much larger change in CO2 levels, such as the ~ 40% rise in CO2 levels since pre-industrial times. There are two parts to this long-term effect. We have what you call “hard data” for one, but not for the other.
The greenhouse hypothesis goes like this:
1. An increase in CO2 or other greenhouse gas increases the ongoing atmospheric absorption of upwelling longwave radiation leaving the surface. Since energy is neither created nor destroyed, that absorbed energy is then radiated by the atmosphere. About half of that radiation goes downwards towards the surface.
2. This small (~1%) increase in surface radiative forcing will then perforce increase the surface temperature.
As I said, we have hard evidence for #1, the increased atmospheric absorption of upwelling radiation. This well-understood effect has a clear basis in physics. It can be measured from satellites, and is accurately emulated by programs like MODTRAN.
On the other hand, we have almost no hard evidence for #2, the response of temperature to a less than !% change in total surface forcing at a time when the system is already in general long-term thermal steady-state (e.g. a variation of global surface temperature of only ± 0.3°C over the 20th century).
I ascribe this to the fact that the temperature of the earth is set by the physics of wind, wave, and water, and not by the forcing. The climate systems responds to such small change by slightly increasing the albedo, and by slightly increasing the surface cooling from thunderstorms and dust devils and other emergent climate phenomena. This response of the climate system is temperature-based and not forcing-based—the various phenomena emerge when temperature exceeds certain limits, not when forcing exceeds certain limits.
And as a result, the system is relatively insensitive to small variations in long-term average forcing.
The surprising fact about the climate is not the change in temperature over the 20th century. The surprise is how tiny that change is, about a tenth of a percent, less than one degree C.
I hope this explains my position.
My best to you, and thanks for your inquiring mind and your interesting post,
w.
Just a check: in talking about small % changes in temperature, you’re doing this as a percentage of the Kelvin temperature, right? Current earth average surface temperature being ~288 K.
Robert Grumbine: I’m using the NOAA database, which is in degrees C, but I’m also using temperature anomalies, and thus deviations from a base value, which could be either K or C and still give the same result. The objective was simply to see whether or not ANY change in monthly atmospheric CO2 concentration could cause ANY change in the curve for monthly temperature anomalies. The absolute value of temperature or CO2 concentration has nothing whatever to do with the analysis.
David: That is a question to Willis.
Willis,
Thanks! It’s nice to have a rational thinker in our midst to offset all this monkey motion that seems to be prevalent nowadays.
BTW, relative to point #2 in your description of AGW, I’ve seen the same thought expressed time and again, but no one seems to have put it up against the second law of thermodynamics, which, of course, states that a radiant body (and that calls up all sorts of irrelevant thoughts, doesn’t it?) can’t add heat to an object that’s as warm, or warmer, than itself. Now, considering the half of thermal radiation that’s absorbed by CO2 and reradiated back at Earth, it can’t possibly be any warmer than when it was emitted by Earth’s surface, and the CO2 has cooled significantly due to the lapse rate, which is likely to make its radiant output still cooler. Therefore, the CO2 blanket that supposedly warms Earth actually can’t do so, and can only slow the rate of heat loss from the planetary surface. By analogy, a blanket over a person can only prevent heat loss from the underlying person, and can raise that person’s temperature only up to a maximum of 98.6 degrees F. It can’t possibly rise above this limit, unless there is some additional heat source under the blanket with the person. Similarly, Earth’s CO2 “blanket” can’t possibly heat the planet in any significant way. Am I missing something here?
yes, you’re missing:
the law of conservation of energy
the second law of thermodynamics (the real one, not the verbal shorthand which requires many caveats that you’ve ignored, and which show you why your description doesn’t apply here)
that blankets work by suppressing convection
that the greenhouse effect (whether CO2 or H2O) works by radiative transfer
that photons don’t have a temperature, merely an energy
then again, you’ve already said you don’t believe photons exist, so the list is longer:
quantum mechanics
start with Planck’s derivation of the black body law (to explain observations). The Theory of Heat Radiation by Max Planck is a good place to start.
Robert Grumbine: Thanks for your input on this!
How, specifically, am I missing the first and second laws of thermodynamics? I invoked the second law in my first sentence above.
Would you please expand on your statement that “blankets work by suppressing convection?” Forgive me, but I had thought that they worked by preventing heat loss, whether by convection or by some other mechanism.
I’m quite well aware that the “greenhouse effect” is supposed to work by radiative transfer, as an even cursory reading of my reply, above, reveals.
Photons and temperature: First, I disagree with A. Einstein that photons are present in EMR (electromagnetic radiation), and therefore I don’t think that EMR contains energy at all. Second, be that as it may, EMR induces resonant vibrations in the molecular bonds of receiving matter (thus generating energy). Hotter sending bodies emit higher-frequency radiation than cooler ones do. In order for there to be a resonant response in receiving matter (in other words, for the EMR to be absorbed), the bonds in that matter must be vibrating at a lower frequency than that of the incident EMR, in other words, the receiving matter must be cooler than the transmitting matter. CO2 in the atmosphere is cooler than Earth’s surface, so no effect should be expected from returning radiation.
You will find an extensive discussion of Planck’s postulate and Planck’s Law, as well as my ideas on the non-existence of photons in EMR (and in space, generally), in my book “The Real World, a Synthesis” on amazon.com.
That you ‘invoke’ something does not mean that you understand it (e.g. also your statements about CO2, buffering, acidity, …) nor does it mean you invoked it correctly. For a starter on thermodynamics, see, for example, Atkins _Physical Chemistry_. It will also help you with your non-understanding of solution chemistry and quantum mechanics.
That you don’t understand how blankets work is yet another reminder that you don’t take Feynman’s admonition seriously.
But the crux of the problems regarding the laws of thermodynamics and more general errors is that you really, really, are ignoring Feynman’s admonition. In rejecting quantum mechanics, you reject most of 20th century physics. Yet you do not first provide explanation of the observations that QM was developed to explain. Start with the ultraviolet catastrophe (Planck), but also include photoelectric effect (Einstein), and spectral lines (Bohr for atoms, Herzberg for molecules). Thence to technologies which can exist only due to QM, including lasers and semiconductors.
In the mean time, you have an overwhelmingly large problem the instant you say radiation does not carry energy. Energy is transferred only by conduction, convection, and radiation (in normal physics, not, apparently, in yours). Since the earth is surrounded by vacuum, ruling out conduction and convection, the only way for it to gain or lose energy is radiation.
In saying radiation does not carry energy, you’re saying that the sun does not warm the earth. Your theory is falsified every day.
[snip]
Robert Grumbine: Your angry, negative tone and ad-hominem remarks indicate to me that further communication with you would be non-productive. I will therefore not take the trouble to respond to you further.
Any anger merely comes from your mirror.
But, by all means, don’t respond. Hide from criticism, which is another part of Feynman’s bending over backwards to avoid fooling yourself. Doesn’t change the fact that you are flagrantly ignoring a century of science. Nor does it change the fact that your theory of radiation prohibits the sun from warming the earth.
Willis: I remain to be convinced that a candle flame can heat Sun. As I understand the situation, absorption in matter occurs by resonance when the vibration of an atomic bond is either 1) of lower intensity but of the same frequency as the incident radiation or 2) of a lower frequency than that of the incident radiation, regardless of intensity. The second law of thermodynamics and common experience holds that a cooler body (candle) cannot transfer heat to a hotter one (Sun). Do you have hard data that refutes either of these statements?
David:
An El Nino event occurs when ENSO sea surface temperatures rise by 0,5 – 2.3 deg.C., and this temperature rise causes average global temperatures of, say,14.5 deg. C. to also rise.
Would this be an example?
Very unlikely.
All ‘bodies’ above absolute zero radiate energy according to the Stefan Bolztmann Law ie to the 4th power of absolute T; a candle will radiate and the sun will radiate.
David Bennett Laing October 20, 2016 at 6:31 pm
David and Robert, I fear that you both are conflating heat and energy. Heat is the NET flow of energy, and can only go from hot to cold.
Radiant energy, on the other hand, always exists in two separate and independent flows, one from A to B, and the other from B to A.
FOR EXAMPLE: If you have a sphere radiating energy at 390 W/m2 and another radiating at 300 W/m2, there is a net HEAT flow of 90 W/m2 between the two, going of course from warmer to cooler.
But this is the net of the two physical flows, one of 390 W/m2 and the other of 300 W/m2.
This means, of course, that when you light a candle in the daytime, the sun ends up warmer than it would without the additional energy flowing from the candle to the sun … crazy, I know, but true. When a photon is emitted by a candle, it knows nothing about the temperature of the object it will hit … and when it is absorbed by that object it hits, it will transfer its energy to that object.
Please take a look at a few of my posts on the subject, linked below.
Best regards to you both,
w.
The Steel Greenhouse 2009-11-17
People Living in Glass Planets 2010-11-27
The R. W. Wood Experiment 2013-02-06
(er, the criticism is part of Feynman’s advice, not the hiding)