Well, I can't give an answer to your question, but I'd love to keep on following this post and find out more. A very important discussion ...

Dear Frank,

I have looked at your graph and i have a couple of questions. First, just an educated guess, but it seems to me that the derivative does not match the trend in the historical period. Second, why the derivative has spikes in the prediction period?

Thanks, I really think this topic is quite relevant, but for me the stated questions should be answered before discussing any further.

Best regards,

   Luis

Thanks Peter,

I think that this discussion is important indeed. There has been speculation on the existence of a runaway mechanism in Global Climate (Change) and here we can observe a rate increase of the rate increase of the atmospheric CO2 mixing ratio as measured at Mauna Loa. I could elicit that by just applying a parametric polynome of the sixth degree, which fits the data perfectly. It is as if in gravity mechanics the gravity constant would increase the closer an object comes to its final resting place on Earth, which your fellow countryman Newton proved is not the case.

Hence, did atmospheric CO2 increase (measured at Mauna Loa) pass the point of no return? If any question about Climate Change is important, then it's that one!

Thanks for your comment. I agree that it not an easy question, that's why I went back to the original data acquired at Mauna Loa, the largest dataset on the CO2 mixing ratio in the Northern hemisphere. I could not help it to fit a polynome through these data. And believe it or not, it is elicited that the first derivative of the change in atmospheric concentration at Mauna Loa is NOT constant at all (green curve in my graph added here in this thread)! According to me this means a strong feedforward mechanism is in play here. But since I am not really a climatologist, I might be kicking into an open door.

Remarkably enough (I am still figgering that out), it seems that the seasonal variation over the years remains rather constant, in contrast with the trend data. Hence, a biological signal surfing on top of an anthropogenic one?

I honestly don't want to speculate on what this phenomenon might mean for our future climate. Let's hope for the best as long as everybody on this planet is still asleep concerning this matter.

Cheers,

Frank

Hi Alonso,

I hope you are doing fine?

Concerning your remarks, you are perfectly right. I also noticed these spikes in the time period of the parametric model, where forecasting starts. The spikes start occuring exactly at the moment where the polynome goes into forecasting mode.

To be honest I find this remarkable, but I have no explanation for this yet, except maybe that data accuracy (polynome coefficients) is in play here and secondly that an incomplete detrending of the measured dataset has occured (due to the former), which is then amplified in the forecasting period of the parametric model. I can check whether an increase in significant digits of the polynome coefficients has an impact on the spikes. The fit of the polynome through the CO2 mixing ratio data at Mauna Loa is close to perfection (see the dotted black line in the graph, representing the polynome and its regression coefficient). I agree that polynome fitting can be tricky as well, but I have taken care not to take too many coefficeints for the polynome (six actually). I don't like bullshit either ;-).

Concerning the observation that the blue curve (the derivative of monthly mean CO2 atmospheric concentration at Maunoa Loa) does not fit the red curve (the actual CO2 mixing ratio's as measured at Mauna Loa. Actually it is not supposed to fit the red curve at all, since it is the first derivative of the red curve. Let's go back to the example of Newtonian mechanics once more. Whenever one has a linearly increasing distance from a starting point, the first derivative of distance in function of time is a constant called zero! Hence it does not fit the distance - time curve at all.

The same goes for the Mauna Loa data, the first derivative of the atmospheric CO2 mixing ratio does not fit the mixing ratio - time curve at all. To be expected as a matter of fact. But I did not expect that the derivative is not constant. It increases sharply in function of time. Newton would be surprised with such a result, indicating a much more complex system than a gravity field. And indeed if there is a really complex system on this Earth then it is the climate system indeed.

How to explain the increase in the first derivative? I can do some backcasting as well, and it is my educated guess that the fiirst derivative will stabilize the further the backcast goes back in time. Logical, because the change in CO2 concentration in the 19th century was much smaller (ice core data) than what is measured now at Mauna Loa.

Hence, I guess we are dealing here with a strong feedforward mechanism in the processes forcing the mixing ratio of atmospheric CO2 at Mauna Loa. I even tend to speculate that the Mauna Loa data are pretty representative for what is taking place in the Northern hemispheres atmosphere. That's just a matter of sucking the CDIAC database dry and spatializing what has been performed with the Mauna Loa data.

So, are we passing the point of no return with respect to Climate Change? Still an open question if you ask me, but the Mauna Loa data do not really comfort me to be honest.

So to answer you specifically I'll do some backcasting, look at accuracies of the polynome coefficients and suck the CDIAC database dry to perform a spatialization. See what comes out of that. Confirmation or not?

Cheers,

Frank

I don't think you can answer this question by fitting higher-order polynomials to the data.  Global policy should be affecting the data in ways that cannot be mathematically predicted.

I noted in about the year 2000 that the data fitted a parabola rather well, and I have used this as a standard to see if emission reductions are having an effect.  And maybe they are, but we have stayed close to this parabola, so increases have tended to balance reductions. 

Dear Frank,

Had read your observation, calculations, conclusions, etc - as well all the other colleques notes.  How to tell - each one is right. So the "climate is changing" as the data and our filling these day with relative higher temp. in comparison to the past.  You think that the rate will remain constant ?  if Not - what will be the new trend ? there are places that the CO2 rate is decreasing ? how all this will effect the life in general ? diversity ? evolution "rates" ? plant cover ? nutrient cycling ? and so on.

The discussion here makes no sense. You can use statistics, polynomial fits etc. to describe sth. but you need physics to explain it. The CO2 is an input variable to the system here and does not increase by itself within the system. In other words, there is no positive feedback and therefore there is no runaway effect at this stage.

If you would observe that CO2 increases more than emitted externally by humans, you might pose your question again. So far, the ocean and to some extent vegetation take up a large fraction of the CO2 increase. A shrinking of this buffering effect should sth. to look at rather than at the absolute CO2 values.

Thanks Kenneth,

I had prepared an answer for Frederik, but it got lost in a Windows crash. I sweared a few times and thought, OK, somewhat later then, that reply for Frederik.

There is still a lot of ground to cover on this topic, I'll show some more graphs later on. Sorry Fredrik, as Kenneth said, genius (sorry for this little change in phrasing) cannot be stuffed back into a bottle ;-). I tend to use it whenever I can.

Cheers,

Frank

Hi Kenneth,

I tried to reproduce your graph with the 1998 spike, but I did not succeed (yet). I calculated the rate of change of the yearly mean CO2 mixing ratio at Mauna Loa.

In the graph added to this message you can observe what I obtained. I plotted the yearly mean rate of change of the Mauna Loa CO2 mixing ratio. I used a polynome of the 4th degree (cannot help it to parameterise the graphs I obtain) and the polynome fits the data perfectly, but actually that's not essential here.

What is essential is that the graph elicits a bending point exactly at 1990 in the yearly mean rate of change of the Mauna Loa CO2 mixing ratio. Might this indicate that starting from 1990 the yearly mean rate of change of the Mauna Loa CO2 mixing ratio goes into second gear? And why? This result looks pretty convincing to me.

Kenneth, how did you produce the graph with the 1998 spike? Did you take the yearly mean rate of change of the Mauna Loa CO2 mixing ratio or the straight yearly mean mixing ratio's? I'll check it out, because I would like to reproduce the 'spike' as well. If the spike corresponds with the El Nino cycle, that would be a big step forward as well.

Thanks guys,

Frank

Frank,

I do not completely understand the question. Do you think that the CO2 concentration at Mauna Loa is growing faster than it could grow under the reported rate of CO2 emissions? I supposed that the rate of CO2 growth is more or less explained in IPCC reports.

Georgii

P.S. It seems to me that your question is partly related to increasing "airborne-fraction" - a scaling factor defined as the ratio of the annual increase in atmospheric CO2 (dCO2/dt) to the annual amount of CO2 emissions from anthropogenic sources (which is not constant). It is normally assumed to be equal to 0.45, but ...

https://en.wikipedia.org/wiki/Airborne_fraction

Hi Georgii,

Good question. I don't think that the CO2 mixing ratio at Mauna Loa is growing faster than the  anthropogenic  input of CO2 in the atmosphere. Mass balances do exist!

However simply put, the atmospheric CO2 is an equilibrium of CO2 input (by human activities) and output by dissolving in oceans and by the photosynthesis of vegetation (phytoplankton and terrestrial vegetation).

When saturation occurs of the dissolution in oceans and/or when vegetation takes up less and less (because the uptake of CO2 by plants saturates with higher CO2 concentrations! And that's not physics but plant physiology!), then the atmospheric CO2 concentration will increase. The increase will be the larger the larger the saturation of photosynthesis and the higher the acidity of our oceans. That's the bio-geophysical model (which some folks in this thread ask for) behind the equilibrium of atmospheric CO2  mixing ratio's. We know what an increase of atmospheric CO2 can bring about in the climate system and especially in local meteorology above terrestrial adn oceanic surfaces. That's applying plain climate physics if you  ask me. Can you dig my point here, the main point being that atmospheric CO2 mixing ratio is determined by a dynamic equilibrium  which shifts according to the magnitude of the CO2 inputs and outputs into the atmosphere. In that respect the residence  time of a molecule of CO2 in the atmosphere (as a rule of thumb) is about a year. Hence the turnover time of this system is rather slack. That's a reason as well why we can observe the seasonal activity of vegetation in the Mauna Loa data as a seasonal ripple on a increasing trend of CO2 mixing ratio. I am kicking in open doors again, because these processes are well-known for some time now.

Cheers,

Frank

Hi Kenneth,

I got the picture of how you produced the spike graph now. I will have a look at the years that El Nino occured to check whether there is a correspondence just once, as you mentioned for the year 1998 or for all El  Nino occurences. Maybe you made that exercise already.

Thanks,

Frank

PS.: Your graph through the 'spike' data, looks pretty much the same as the one I produced when plotting the yearly mean rate of change of the Mauna Loa CO2 mixing ratio's. 1990 seems to be the bending point as well. Did you take the 4th order poolynomial with the same values for its coefficients as the one I mention here above? Would be good to compare the coefficients.

Hi Georgii,

It's simple, I just plotted the first derivative of the Monthly mean Mauna Loa CO2 mixing ratio's. That's all. You have seen the result in my first graph. I did not divide the first derivative by emissions. By the way, who on Earth has exact data on global CO2 emissions? Many countries do not report these data. Hence calculating an airborne fraction as defined in your thread would give extremely unreliable results which would learn us what? That the global emission inventory is hopelessly incomplete? Don't we know that since years? I even think that some countries consider their CO2 emissions as a National secret!

That Georgii does not mean that the airborne fraction is not an interesting variable. On the contrary!

Cheers,

Frank

Frank,

You are talking about feed forward control but that cannot lead to a runaway situation. What we need to fear is positive feedback.  When the positive feedback is greater than 100% then a runaway situation occurs and there is an instantaneous change.

For instance, the increase in carbon dioxide is causing the permafrost to melt giving off methane, a more powerful greenhouse gas.  The methane will then cause more melting and more methane, in a positive feedback loop. That process may continue slowly or increase out of control in a runaway.

As well as runaway situations there are also tipping points. That is when the methane being released is greater than that being oxidised by the atmosophere, so that any increase in temperature (e.g. due to CO2) will lead to further heating.

The situation with CO2 is that at present about a third of that being emitted by us is being dissolved in the oceans and another third is absorbed on land. But if the global temperature rises enough then the oceans will become a net emitter of CO2. See Kenneth's chart of CO2 during El Ninos. But we do not know how much CO2 we have to add to the atmosphere to pass that tipping point.

HTH.

Cheers, Alastair.

Thanks, Alastair, I completely agree.

And there are indeed more of such possible positive feedbacks such as greenhouse gases captured in ice being released in warming and temperature related ocean floor releases of such gases. The positive feedbacks are likely to be more worrisome  than other processes contributing as discussed.

Cheers, Kees

Hi Kees,

It was the release of methane from the Arctic sea floor which was at the back of my mind. It is described here in a FEEM Lecture by Professor Peter Wadhams: http://www.feem.it/getpage.aspx?id=7550&sez=For&padre=22&idsub=108

We have already passed the tipping point for the removal of Arctic sea ice. Since the ice albedo is a positive feedback then the ice may disappear abruptly as the ice albedo or the release of methane leads to a runaway. The consequent change in albedo may lead to a runaway warming in the Northern Hemisphere as happened at the start of the Holocene. http://press.princeton.edu/titles/6916.html

We may also have passed the tipping point for the complete melting of the Greenland ice sheet. It is already recording an accelerating loss of ice which is unstoppable since when it melts the altitude of the upper surface falls to heights where the air temperature is greater.

Cheers, Alastair.

Hi Alastair,

Are you sure about the magnitude of global emissions. Do you think that for example China, India, Bangla-Desh the majority of African countries, Russia or North-Korea and many other countries, produce reliable emission date? In some of these countries emissions of CO2 are even kept a secret if these countries even make inventories at all ! I have the  strongest doubts on that!

That's why I leave out global ermission estimates, because they are - according to me - utterly unreliable, at the global scale, I think even their magnitude is underestimated severely. A small minority of countries does a good job though, but we are dealing with a global phenomenon here. Hence?

in the interpretation of Mauna Loa data - because that's what we are dealing with here, how do you corroborate the one third  - one third rule for CO2 absorption in oceans and terrestrial vegetation. This would mean that 60%  of all CO2 emissions would be taken up again.Well the terrestrial science community  and remote sensers are still trying to accurately pin down the uptake of CO2 by vegetation. The same for the oceanographers and CO2 dissolution in oceans and the subsequent uptake by phtytoplankton, and that with oceans which acidify?  CO2 oceanography is a field of research which is even more complicated than the terrestrial one, especially to get a grasp on the CO2 uptake figures by those systems. I think you are strongly underestimating the complexity of these systems Alastair. Hence, Are you not speculating a little but too much collaegue and introducing some very crude assumptions and rules of thumb?

Figures and facts that's what we need, not the mumbo jumbo of rules of thumb. And what is the difference between a runaway process and a process with a strong positive feedback? Semantics or physics? Please man!

CHeers,

Frank

Frank,

When I first became interested in these matters, the problem was where is the missing carbon sink? It has now been found, see

http://www.nature.com/climate/2007/0708/full/climate.2007.35.html

and references there in.

In 2007 she writes:

"Of the 8 billion tonnes of carbon that human activity produces each year — 6.4 from fossil-fuel emissions, and 1.6 from deforestation, mainly in the tropics1 — on average, 3.2 billion tonnes remain in the atmosphere, 2.2 billion tonnes are stored in the oceans and 2.6 billion tonnes are sucked up by land-based carbon sinks, mainly forests. "

So my estimate of 1/3 each is roughly correct, but of course those figures will be different now, and as you say how accurate are they? However, I do not believe that they were drawn up by scientists who were as naive as you suggest.

Kenneth,

You are obviously unaware of this open access paper by  Francis and Varvus (2015) "Evidence for a wavier jet stream in response to rapid Arctic warming" http://iopscience.iop.org/1748-9326/10/1/014005

Changes in the jet stream appear to be a result of the melting Arctic sea ice, which I am sure you will agree is caused by the increase in global carbon dioxide.

Frank,

You need not national reports to calculate the world total amount of CO2 emissions from fossil fuels. They are basically proportional to global supply of coal, gas, and oil. I do not think  that such data could be easily kept in secret from global markets. Anyway, there is completely non-governmental way to check the data on global CO2 emissions growth: the C13 measurements. You may download these data from the same website as Mauna Loa CO2 data.

As to the growing dCO2/dt, it could be constant only in the case if the difference between the annual amount of CO2 emissions and the annual magnitude of global net carbon uptake were constant.  However, it is not constant,  this difference is proportional to the annual amount of emissions. Since annual amount of emissions is growing, it is not surprising that the first derivative of [CO2] is growing too.

Best wishes,

Georgii

 In science, if we want to investigate any scientific problem, major influence factors should be given more priority to be investigated. However, only one field is an exception: that is global climate warming. Many scientists ignore the major climate influence factors and consider greenhouse effect  due to human reasons to be the main reason , although they knew it is much smaller than the major climate factors.

     Their logic is that the recent unusual trend of land surface temperature is non-periodic, major periodic factors only cause periodic changes.

     Recently, after more than a half year's effort, we prove that the monthly anomaly of global land surface temperature can be fitted perfectly by a group of periodic functions, the verification results indicate that the correlated coefficients of more than 15,000 periodic functions are all more than 0.9.(Last year, I mentioned I found a function which its the correlated coefficient is 0.8)

    I send the paper to one of the toppest IF magazines, the editor said" we know  your reseach result is absolutely right,but many people already believe in this view, thousands of papers have already published , so we have to reject your paper. And we also tell you that  no magazine is willing to publish your paper".

    One anonymous email is send to me with the following word :

         "Birth is much, but breeding is more."

 In science, if we want to investigate any scientific problem, major influence factors should be given more priority to be investigated. However, only one field is an exception: that is global climate warming. Many scientists ignore the major climate influence factors and consider greenhouse effect  due to human reasons to be the main reason , although they knew it is much smaller than the major climate factors.

     Their logic is that the recent unusual trend of land surface temperature is non-periodic, major periodic factors only cause periodic changes.

     Recently, after more than a half year's effort, we prove that the monthly anomaly of global land surface temperature can be fitted perfectly by a group of periodic functions, the verification results indicate that the correlated coefficients of more than 15,000 periodic functions are all more than 0.9.(Last year, I mentioned I found a function which its the correlated coefficient is 0.8)

    I send the paper to one of the toppest IF magazines, the editor said" we know  your reseach result is absolutely right,but many people already believe in this view, thousands of papers have already published , so we have to reject your paper. And we also tell you that  no magazine is willing to publish your paper".

    One anonymous email is send to me with the following word :

         "Birth is much, but breeding is more."

Kenneth,

You are quite correct!

Francis et al. and I not only assume that CO2 is melting the Arctic se ice, we are also convinced that it is true, whereas you appear to be convinced otherwise. It appears that here in no common groung between us :-(

For those who care to dispute that the Arctic se ice is melting, then my web page at:

http://www.abmcdonald.freeserve.co.uk/north.htm

gives a graphic view of the ice decline over the last 17 years.

Kenneth,

The sea ice retreat in the 30's  may not have been natural! CO2 had been increasing since the start of the Industrial Revolution, enabling Callendar(1938)  to be the first to observe global warming.

There is a chart here: http://www.scarlet-jade.com/wp-content/uploads/2015/07/Global-Temperature-Anomalies-to-2015_06.png prepared from the GISS data at http://data.giss.nasa.gov/gistemp/tabledata_v3/GLB.Ts+dSST.txt

It shows quite clearly the temperature rise upto 1940 followed by the rise from 1970 until today.

Ken,

The NASA charts are anomalies as you say but at the end of the file it explains that to convert to absolute temperature you add 14.40 C. They show a rise after 1995.

According to NASA the average annual temperature from Jan to Nov in 1995 was 14.86C and  15.14C in 2014 for Global Land-Sea. Are you looking at the contiguous US records?

Ken,

You wrote "

What was the CO2 in the 19320s? Or, during the Viking era?"

AFAIK, It was 230 ppm in the 1930s and 280 ppm in the Viking era.

Then you asked "Have you checked to see how much Callendar's selected cities have increased since 1938?" No I have not. I have better things to do with my time!

You continued "On average they have not...but it's not statistically meaningful. Those that are warmer are subject to urban heat increases since then." If you knew the answer why ask me?

"Remember, that 1938 was also a very warm year that created an "extreme" hurricane as strong as "super storm" Sandy...and it hit the same place! "

That is what I am saying, global warming began around 1910. See the GISS chart.

Ken,

You wrote "Still no answer to my questions. Why?" I have answered your questions now, but the reason I did not do it earlier is I have more important workd to do.

It is true that in 1996, the year 1995 was the hottest ever recorded but since then other hyears have been hotter e.g. 2014. The pea does keep moving as increasing levels of CO2 drive temperatures ever higher. But not monotomically - the climate is chaotic. Just like a tennis ball dropped at the top of the stairs, it will rise as well as fall as it bounces down, but the final result is that it lands at the bottom, just as the final result of increasing CO2 will be a runaway rise.

If you don't believe me that temperatures have risen since 1995 read this press report only a few months old:

http://www.washingtonpost.com/news/energy-environment/wp/2015/01/23/sorry-skeptics-nasa-and-noaa-were-right-about-the-2014-temperature-record/

Cheers, Alastair.

Ken,

You are comparing the NASA data for 2014 with the CRU data for 1995. The CRU data does not include the Arctic, which means their base temperature is higher, which explains the higher 1995 value. It also explains their hiatus, since most of the warming has been in the Arctic from where the melt water has escaped to the North Atlantic holding their global temperature lower.

This has been useful after all :-)

Cheers, Alastair.

No I can't comprehend what your argument is! It seems to be a set of questions that you know the answer to, but want me to investigate. Sorry don't have the time. Here is the latest HadCRUT data:

http://www.cru.uea.ac.uk/cru/data/temperature/HadCRUT4.pdf

2014 clearly comes out as warmest, except in the SH.

Cherio, Alastair.

Hi Kenneth and colleagues,

I promised to look into the relationship between the yearly change of atmospheric CO2 concentration (ppmv) and the mean yearly ENSO anomaly. Would there be a relationship between both? Hawaii is positioned not so bad to pick up an impact of ENSO on changes in atmospheric CO2 concentration. Hence?

Well here is my short study on the relationship between detrended yearly change of atmospheric CO2 concentration at Mauna Loa and the yearly average ENSO anomaly.

The correspondence according to me is striking. The regression is noisy, but is very close to a 1:1 relationship. Hence ENSO can be a predictor for detrended yearly atmospheric CO2 concentrations measured at Mauna Loa. according to me we can observe a clearcut coupling between a periodic (7 to 9 year cycle) oceanic phenomenon and the atmospheric CO2 concentration as measured at Mauna Loa after detrending.

Anyone wants to explain this? Anyone has observed this in some paper? Or am I kicking into open doors again? Of course this might a specific phenomenon for the Central Pacific. Has to be checked out with other observations in the Atlantic somwhere close the West-African coast.

Look at the volcano eruptions in the last figure! Coincidence?

Cheers folks, something new to figger out again?

Eagerly awaiting your comments.

Frank

Hi Frank,

I now understand ("comprehend") what you were saying in your original findings. I would phrase it as carbon dioxide is increasing exponentially, driven by feed forward anthropogenic emissions. That is similar to a runaway state. It happens if the feed forward forcing is driven by a positive feedback. In this case population growth  produces the positive feedback. The more people you have then the more children and hence a little later more people. But there is also economic growth. In order to sustain that we are burning more and more fossil fuels each year. Perhaps this second financial crisis will put an end to that. The previous one did slow growth temporarily. Or the upcoming Paris Conference may result in a cut back in CO2 emissions. Let's hope so.

There are two main natural sources of CO2 for the atmosphere: the oceans and the land. The solubility of carbon dioxide is inversely proportional to temperature. As the SST increases more CO2 enters the atmosphere. Since the oceans are dominant in the SH there is a rise in CO2 in the Austral summer when the SST rises there. The other source of CO2 is respiration and decomposition of vegetation on land, which peaks during the Boreal spring. If you analyse the monthly, rather than annual, CO2 records I think you will find that there are two peaks each year in CO2 concentration.

El Nino also raises the SST, so it is not surprising that CO2 increases when they happen. I would expect that you will find that during the Austal summer corresponding to an El Nino peaks even more than usual. It will be interesting to hear your conclusions.

Hi Kenneth,

I agree, the comparison between the ENSO SST yearly anomalies and the detrended Mauna Loa atmospheric CO2 yearly difference data is enigmatic. But concerning the correlation graph between those two variables, I conclude that

1) The correlation coefficient is rather low (R²=0.3), but,...

2) the linear regression curve is strikingly accurate, meaning that though the point spread is large, the regression curve goes straight through the origin (intercept is 0.025) and deviates only by a mere 2 % from a 1:1 relationship, meaning a regression curve with a slope of 1. That's pretty exceptional indeed and not to be thrown in the garbage can.

Maybe we can see a bigger picture here. Mauna Loa is located on Hawaii. Hawaii is located pretty well in the midpoint of the 'circle of fire', meaning, it is surrounded by a circle of volcanoes and subduction zones (with earthquakes) from Japan till the South of Chili. The ring of fire is shaped like a horseshoe (see figure). That's why I suggested to have a look at the Atlantic Ocean dynamics and take a CDIAC CO2 measuring stationsdata which is NOT surrounded by a 'circle of fire' as the one around the Pacific. Clearly we can see the impacts of volcanic eruptions (see my previous figures on the ENSO anomaly as well as the Mauna Loa yearly  atmospheric CO2 differences. The strongest ENSO anomalies seem to be coinciding with volcanic eruptions except for the 1997 anomaly, which is the largest anomaly of them all.

I looked further and found that a large eruption took place on the 25th of June 1997 in the Souffriere Hills on the Isle of Montserrat in the Carribean. It resulted in the deaths of nineteen people. The island's airport was directly in the path of the main pyroclastic flow and was completely destroyed. Montserrat's tourist industry was also destroyed. Montserrat is located just  North of Guadeloupe and  a bit South from Puerto Rico. It is located 16 ° N of the eqquator, but easterly from the 'ring of fire', but in the wake of the ENSO impact on climate (and Atlantic SST?). Are the ENSO anomalies triggered by volcanic eruptions? Maybe, but Mauna Loa also follows the dynamics of the ENSO anomalies quite closely as you can see in the graphs I made. Maybe the Carribean plays a role in this puzzle too. Maybe smaller eruptions are the cause of both the ENSO anomaly dynamics and the Mauna Loa yearly CO2 difference dynamics. Let's not forget that the Pacific contains two systems, which have an influence on the Mauna Loa yearly CO2 differences.

One is its phytoplankton, which determines the uptake of CO2. Secondly, the equatorial Pacific SST determines the dissolution and release of CO2 in the Pacific Ocean. Dissolution decreases with increasing SST (see the graph). Hence with each strong ENSO less CO2 is dissolved in the pacific and the CO2 difference at Mauna Loa will increase once more. Moreover when less CO2 is disolved in the pacific, phytoplankton will have less bicarbonate to take up and hence CO2 uptake by phytoplankton in the equatorial Pacific is reduced even more, increasing the difference of atmospheric CO2 as measured at Mauna Loa.

Last but not least the atmosphere itself, due to volcanic eruptions, will be regularly more loaded with aerosols and hence its optical density (OD550) will increase. This results in less available Photosynthetic Active Radiation (PAR) for phytoplankton absorbed (for photosynthesis) and hence an aditional decrease in CO2 uptake will take place and a concomittant increase in yearly dfifference in atmospheric CO2 at Mauna Loa will be the result. Do not ask me whether the impacts I cited are synergistic or additional. But according to me, these are the processes responsible for the Mauna Loa inter-annual dynamics in atmospheric CO2. 

In short, the activity of the Ring of Fire determines the dynamics of the Mauna Loa yearly atmospheric CO2 differences as well as the ENSO anomalies. All but 3 of the world's 25 largest volcanic eruptions of the last 11,700 years occurred at volcanoes in the Ring of Fire. Check it out!

Intruiging will be to investigate whether the Atlantic without a vast ring of fire like  the pacific shows the same behaviour as at Mauna Loa. At least it is known that the Northern Atlantic does NOT elicit ENSO anomaly patterns and what's more, the Northern part of this Ocean does take up CO2, by dissolution and phytoplankton activity. The Pacific during a la Nina releases it. Even more interesting is to know the influence of terrestrial vegetation on the CO2 signal. This might be looked at in the Southern hemisphere CDIAC stations, which are remote from vast continental areas (with vegtation in the tropics). As far as I know, the monthly CO2 differences in the oceans of the Southern hemisphere do not show a seasonal signal like the one I have shown for Mauna Loa on top of an increasing trend of the CO2 mixing ratio. I'll try to give an illustration of this phenomenon soon.

Another question is, whether the ring of fire shows the same periodicity of 7 to 9 years as the ENSO SST anomalies? One would be inclined to suppose so. Maybe the answer can be found in the dynamics of the continental plate shifts, and the pressure build-up and accompanying volcanic eruptions and earthquakes.To be checked I would say. It might corroborate my hypthesis.

I think we are closing in on a logical explanation for the dynamics of the Mauna Loa atmospheric CO2 yearly differences. But the puzzle is not completely disentangled yet. Alfred Wegener took his time too, to prove his continental drift hypothesis.

Cheers,

Frank