UPDATE: I received a reply tonight from Pieter Tans, who is the manager for the MLO data, it is another error in presenting the data, similar to what happened with GISS in October, a monthly data value was carried over. In this case, November to December. – Anthony
From: “Pieter Tans” <Pieter.Tans@xxxxx.xxx>
Sent: Sunday, January 11, 2009 7:24 PM
To: “Anthony Watts ” <awatts@xxxxx.xxx>
Subject: Re: Questions on currently posted 2008 MLO data
> Anthony,
>
> The posted December figure is an error. It will probably be fixed
> tomorrow. The error does not appear on my computer. Our web site is
> run by a separate server dedicated to communicate outside the firewall.
> At this moment I don’t know why it repeated the November value for December.
>
> Sorry about this mishap.
>
> Pieter Tans
The year end CO2 data for the Mauna Loa Observatory is out, and it shows that the trend of Co2 increase has slowed. This year saw the lowest increase in the annual mean growth rate ever in the Mauna Loa Co2 Record: 0.24 parts per million.
Whether this is real, a data error, or something else remains to be seen. As we’ve learned previously, the Mauna Loa record is not infallible and can be adjusted post facto. To MLO’s credit, they have been responsive to queries from myself and others, and have pledged to make improvements to the process. They now have a change log, but there is no mention of the December 2008 data in it.
Here is the graph recently posted by MLO. Notice the two dips in 2008.
The blue line represents the mean value, while the red line is the monthly values. Note that the red line shows seasonal variance related to earth’s own processes that emit and absorb CO2. In the case of the 2008 value of 0.24 ppm/yr it comes on the heels of 2007’s strong year of 2.14 ppm/yr which by itself isn’t that remarkable, being only the seventh highest year in the record.
What is interesting though is the correlation of lower CO2 to a cooler 2008, suggesting that natural mechanisms, particularly the oceans, played a role in the the lower Co2 value for 2008. There are also other likely drivers of this change. For the layman reader, this is essentially the “soda pop effect”. As anyone knows, warm soda pop tends to ‘fizz’ vigorously, while cold soda pop is more tame. This is because colder water can absorb more Co2 than warmer water, and warmer water releases it more easily, especially when agitated. Lesson here, and citing from experience; don’t leave a 12 pack of Coke in your car on a hot summer day. 😉
Here is a graph of Carbon Dioxide solubility in water versus temperature:
Here is the entire annual mean growth rate MLO data set:
year ppm/yr
1959 0.95
1960 0.51
1961 0.95
1962 0.69
1963 0.73
1964 0.29
1965 0.98
1966 1.23
1967 0.75
1968 1.02
1969 1.34
1970 1.02
1971 0.82
1972 1.76
1973 1.18
1974 0.78
1975 1.10
1976 0.91
1977 2.09
1978 1.31
1979 1.68
1980 1.80
1981 1.43
1982 0.72
1983 2.16
1984 1.37
1985 1.24
1986 1.51
1987 2.33
1988 2.09
1989 1.27
1990 1.31
1991 1.02
1992 0.43
1993 1.35
1994 1.90
1995 1.98
1996 1.19
1997 1.96
1998 2.93
1999 0.94
2000 1.74
2001 1.59
2002 2.56
2003 2.25
2004 1.62
2005 2.53
2006 1.72
2007 2.14
2008 0.24
Here a copy of the CO2 values of the last three months:
| Month | Mean | Interpolated | Trend(seasonally corrected) |
| 2008 10 | 382.98 | 382.98 | 386.34 |
| 2008 11 | 384.11 | 384.11 | 386.19 |
| 2008 12 | 384.11 | 384.11 | 385.03 |
Source data from MLO is here
Note the identical months of November and December. It could be a GISS October2008 kind of carryover error, it could also be real. The global values for December 2008 are not yet out. Mauna Loa is only one of many CO2 reporting stations.
If the data is real, there is a dead stop in the monthly numbers, which results, when seasonally corrected, in a considerable decrease, not seen in previous Decembers through the entire record.
As MLO points out:
“The last year of data are still preliminary, pending recalibrations of reference gases and other quality control checks.”
As I previously mentioned, some reasons could be cooling of oceans. In particular the Pacific where we’ve had a La Nina event. See this guest post from Dr. Roy Spencer on how the oceans could be driving the observed Co2 changes. The other possibility is the global economic crisis. This has led to lowered consumption of fossil fuels, particularly gasoline, which saw a significant drop in miles driven this past year due to high prices and other economic uncertainties.
Most probably it is a combination of events or possibly an error. Stay tuned.
h/t to Werner Weber and many other people who notified me
sekerob, your delta mlo is different from cdiac?
isnt the 1999 delta 1.45 for MLO and 1.63 for world?
from the noaa data for annual means with delta
ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_annmean_mlo.txt
1998 366.5 3.03
1999 368.14 1.64
2000 369.4 1.26
yet your data is the noaa delta data!
ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_gr_mlo.txt
this is quite odd: .99 and 1.64 from supposedly the same data
should be .94 vs 1.64 not .99 vs 1.64
Annualized Mauna Loa dCO2/dt averaged ~1ppm/year from 1958 to ~1978, then ~1.5 ppm/year from ~1978 to ~2001, then >2ppm/year from ~2001 to ~2006, and since then has dropped below 2ppm/year (consistent with strong global cooling since January 2007).
However humanmade CO2 emissions have continued to increase over the past few years, as they have every year over the past century or more. Why then is atmospheric dCO2/dt not also increasing?
Mauna Loa (and global) dCO2/dt correlates well with the Lower Troposphere temperature anomaly, but as I noted in my January 2008 paper*, CO2 lags temperature by ~9 months.
The impact of global temperature on atmospheric CO2 concentrations is apparent.
The impact of atmospheric CO2 concentration on global temperature is much more difficult to demonstrate, probably because it is insignificant.
Regards, Allan
_________________________
Annualized Mauna Loa dCO2/dt has “gone negative” a few times in the past (calculating dCO2/dt from monthly data, by taking CO2MonthX (year n+1) minus CO2MonthX (year n) to minimize the seasonal CO2 “sawtooth”.)
These 12-month periods are (Year-Month ending):
1959-8
1963-9
1964-5
1965-1
1965-5
1965-6
1971-4
1974-6
1974-8
1974-9
Has this not happened recently because of increased humanmade CO2 emissions, or because the world has, until recently, been getting warmer?
I noted in a paper published one year ago that dCO2/dt changes contemporaneously with “average” global temperature, and CO2 lags temperature by ~9 months.
* For those who are interested, my paper and spreadsheet are at:
http://icecap.us/index.php/go/joes-blog/carbon_dioxide_in_not_the_primary_cause_of_global_warming_the_future_can_no/
CO2 data from Mauna Loa:
ftp://ftp.cmdl.noaa.gov/ccg/co2/trends/co2_mm_mlo.txt
“With the amount of unemployed people going up tremendously and manufacturers shutting down plants worldwide, one would think that energy consumption, miles driven, and anthropogenic CO2 emissions would be all be dropping like a rock.”
I think it taken as a percentage of the whole?…it would be insignificant. Number of plants shut down vs. number still running and new plants coming online?…don’t know the number, but I’m sure it’s much smaller than the MSM would lead us to believe. Likewise with unemployed people…remember, it averages %6. And for many of those, they may be driving as much or more as they were when they were employed. Not all, but some. So again…overall impact is likely negligible.
“What is intriguing to me is the Detroit Auto Show currently underway. Lots of hybrids and electric vehicles there. Should have a measureable downward impact on U.S. fuel consumption for the next few years. ”
What is the percentage of man-made C02 released by U.S. automobiles?…again…I’m betting it’s a very small percentage.
JimB
Pamela Gray (22:48:28) :
Hi Pamela
It is AIRS data you must be talking about: They have published but it is for around 10000 meters, I think. They cannot go below 5000. http://airs.jpl.nasa.gov/
Have a look at the animations
http://svs.gsfc.nasa.gov/vis/a000000/a003500/a003562/
The new satellite will go up in February, will be able to read from the ground up, but depends on sunlight so no nightly CO2, except by modelling daytime data. It is called OCO
oco.jpl.nasa.gov the pr links or the right are more informative.
Pamela:
“If it creates a mild acid-based carbonate solution, no wonder dripping water off a gutter can drill a hole into concrete below.
hmmmm”
How do we know how much, if any, of the hole is caused by the acidic solution, and how much is caused by the force of a drop of water slamming into it from a height of 8-10′?
;*)
JimB
At the suggestion of Les Johnson I made some changes to the annual rate of CO2 addition graphs at My Climate Buzz: Trends in Yearly CO2 Increase
Hopefully they will now present a clearer or more accurate picture. I also added a couple comparing the trends in different ways. I did attempt to discuss (defend) the graphs and concept at Tomino’s site. It seems that if you have a differing opinion then an ad-hominem outburst is soon to be delivered. That reminds me of a doctor who had no patience for nor wanted to hear his nurse’s opinion. He went ahead and amputated the wrong leg from his patient despite her objections.
“Because a fully mature forest does not remove CO2 from the air.”
Another issue with Mauna Loa, Pamela’s ‘stairstep to heaven’. Keeling and minions believe it biogenic owing to terrestrial plants whose center of mass would lie around Tropic of Cancer.
But the peak of sequestering by the data is October. It cannot follow July/August.
Now consider the SO as the 800 pound gorilla. Remember 7% more insolation.
George E.Smith:
It would appear to me from the missing three-D graph, that the southern ocean take up of CO2 is pretty much constant 12 months out of the year with almost no cyclic change at all, but the increased and still increasing bio-mass in the northern hemisphere accounts for most of the cyclic variation throughout the year.
Look at the links I gave to Pamela, above, to see how the planet breathes out CO2 in spring , a CO2 that is not homogeneous at all.
If you go to Lucy’s links you will find a running discussion about CO2 measurements and the cherry picking that has been going on,
http://www.greenworldtrust.org.uk/Forum/phpBB2/viewforum.php?f=22
It seems that the Keeling legacy is to choose the most stable time of day and region to measure CO2, that is why the locations are remote and on mountains.
I have been challenging this, because it seems to me that it is as if we would decide to measure temperatures where they are stable and away from heat sources, like the sun. i.e. at night and in remote places !!
I think this is because they have assumed a priori that CO2 is well and easily mixed in the atmosphere, which the AIRS data show that it is not.
Christian Bultmann:
Thanks for the link, it affirms and corrects my thinking. On the latter, the cold trap error seems to be similar in size to the H2SO4 dessication method used in French volumetric methods bridging the ice core data to WWII era.
Gary Palmgren (19:43:09)
Thanks for that interesting PDF link, which shows pretty conclusively that the persistence of atmospheric CO2 is only about 12.9 years. Another nail in the AGW coffin.
Gary Palmgren (19:43:09)
Thanks for that interesting PDF link, which shows pretty conclusively that the persistence of atmospheric CO2 is only about 12.9 years. Another nail in the AGW coffin.
You misunderstand what that 12.9 figure represents, it is the lifetime of an individual molecule in the atmosphere not the lifetime of a perturbation in the CO2 concentration. The measurement effectively involved seeding the atmospheric CO2 and observing the exchange with the ocean. Even in equilibrium between the atmosphere and ocean seeded CO2 will enter the ocean and be replaced by a CO2 molecule from the ocean, the concentration of CO2 in both ocean and atmosphere remains the same, but the concentration of seed in the atmosphere will go down and in the ocean will increase. The removal of the excess CO2 should the emission of CO2 fall below the capacity of the biosphere/upper ocean then the [CO2] would drop slowly to a new equilibrium over ~century (depending on the annual deficit).
“The removal of the excess CO2 should the emission of CO2 fall below the capacity of the biosphere/upper ocean then the [CO2] would drop slowly to a new equilibrium”
A fantasy you have no means to demonstrate. Spencer, here at WUWT, had a post demonstrating that the variance in 13C/12C of the MLO seasonal cycle and long-term trends (on which the seasonal rides) were identical under F test.
Now the trend is the sum of all fluences interfacing with the atmosphere where the seasonal is some undetermined subset with recurrent effect.
Now if the rise in the trend is due to the anthropogenic flux the variances would differ markedly as the contributions of the various fluences vary by orders of magnitude, e.g., the biogenic flux is roughly 2 orders greater than the anthropogenic. There is no conceivable way they would march in lockstep both seasonally and over time.
Is it really possible that there is no CO2 contamination of these samples from the Volcanoes?
There are many folks here much more knowledgeable than I am. Is there an explanation for the layman that describes how one can be certain of having no contamination from local sources? Not only the volcano that the monitoring station sits on, but also the longest continually erupting volcano on earth about 15 miles away. Not to mention the development of the Big Island since these measurements began.
How is a lack of contamination determined? How can one know?
Gary Palmgren: Seriously, if the lifetime of all molecules combined were only 12.9 years, don’t you think it would have started to equilibrate to our CO2 emissions and ocean warming by now. The effective lifetime is on the order of centuries (and I’m a skeptic!).
Lee Kington. Your graph clearly shows the 1998 peak, which demonstrates the boiling up of CO2 from sea waters due to 1998 el Nino high temps.
“Seriously, if the lifetime of all molecules combined were only 12.9 years, don’t you think it would have started to equilibrate to our CO2 emissions and ocean warming by now.”
No idea where the force of this argument lies.
The production of 14C is high in the stratosphere at high latitudes. A peak in production precedes the peak in life forms by roughly 60 years (de Jager, Usoskin 2004). Considering the poorly mixed nature of CO2 this upper limit seems consistent with an average residence time of 13 years to me.
Gary: Yes, I completely agree with you, but the average residence time doesn’t matter as much as you think (it is important, but not as limiting as many here are implying). We are adding molecules with a *mean* residence of 12.9 years. Some have a residence time of 5 seconds and some of them have a residence time of 10000 years (this is why CO2 is so complicated). If we emit fossil fuels, all of these molecules (5s, 13y or 10000y) are molecules that would not otherwise have been there in the first place, and effect is felt for a lot longer than the 12.9 years. Why? Firstly, some of the molecules are around, by low probability chance, for hundreds of years or more. Secondly, these molecules will compete with naturally produced CO2 for uptake). I am a total climate change skeptic for a wide variety of reasons, and I don’t think CO2 is a problem at all, but I *do* think it is reasonable to propose that a long-term concentration increase results from emissions from human activities and the oceans.
lulo and Phil.
Beck (papers at Icecap) reported early direct measurement of CO2 by chemical means the value by 1820 rising to 450ppm which I take to follow Tambora in 1815 and other smaller but large eruptions in 1812 & 1814. Tambora ejected 100 Km^3, 20% of which was H2O and CO2 to support the ultraplinian column extending into the Mesosphere.
The level took between one and two decades to return to ~300ppm, or the average for the period.
So by two empirical tests, one unrepeatable on command, your notion fails.
“” DocMartyn (16:52:04) :
Have a warm damp mass of air filled with 390 ppm CO2 and cool it . The pressure drops and water in liquid form condenses. It absorbs CO2. Absorbing CO2 lower the freezing point of water cools and absorbs more CO2. The water turns to snow and lands in the ocean. Co2 is scrubbed from the air. “”
Thanks Docmartin. I didn’t mean to create an impression that I had no idea what removal processes are in play. I just said whatever processes they were, they can sure remove a lot of CO2 fast.
The water scrubbing of course can easily occur at other places than the arctic. True the colder the water, the more CO2 it can contain, but rain, sleet and snow can all form well out of the arctic,a nd remove a lot of CO2.
So why don’t we see a massive removal in the southern oceans. If the process you describe requires open water to take place, why wouldn’t it happen continuously in the southern ocean instead of cyclically as at the North Pole.
I must say I found it interesting to learn that the north pole becomes ice free every year; I would have thought that there would be more or less permanent ice there for a large part of the arctic ocean.
As for the wide availability of the NOAA 3-D curve; I have numerous hard copies on dead tree that I luckily made when it was on their site; but I can enter the exact same URL including the exact letter cases, and it does not bring up the curve.
Maybe somebody else can popit up from: http://www.mlo.noaa.gov/Projects/GASES/co2glob.htm
Which is where my copy comes from
I’ve seen those papers which show CO2 decay curves with long time constant tails some of them thousands or millions of years assignable to rock weathering and the like. Not impressed. No I didn’t say those processes don’t occur, I had never thought about such things but I believe it if a geologist says that rocks ougas CO2 due to weathering.
Still not impressed. We are told, that the long term steady state atmospheric abundance of CO2 was about 280 ppm; which is about the lowest that CO2 has ever been in the last billion years or so; so even at 385 ppm, we are enjoying some of the lowest CO2 levels this planet has ever had.
I’ve not seen one shred of evidence, that establishes that any of those long time constant tails, even the ones with 200 year “residence times” kicks in at CO2 levels as high as even 280 ppm.
So I have seen no evidence that all of the excess over 280 ppm, wouldn’t disappear in something of the order of the time frame I alluded to. I wasn’t aware of the 12.9 year computation based on nuke test aftermath, but given that actual measured result, I consider my simple stick on a sandy beach estimate of 9 years to be not too shabby. Given that everything in climate science modelling has a mandatory 3:1 fudge factor in predicted outcome. that puts my 9 year number fair and square in the middle of a 5-15 year fudge window, so 12.9 years is close enough for me.
I’m astonished by Gary Gulrud’s input that in 1820 the abundance was 450 ppm.
Michael Mann seems to insist that the earth temperature was constant for several thousand years prior to his hockey stick foot, and therefore CO2 should have been constant too, since supposedly one causes the other and the order hardly matters much if both are constant.
Are we saying that the 280 ppm long term stable value is a myth?
“Phil. (08:49:01) :
You misunderstand what that 12.9 figure represents, it is the lifetime of an individual molecule in the atmosphere not the lifetime of a perturbation in the CO2 concentration. The measurement effectively involved seeding the atmospheric CO2 and observing the exchange with the ocean. Even in equilibrium between the atmosphere and ocean seeded CO2 will enter the ocean and be replaced by a CO2 molecule from the ocean, the concentration of CO2 in both ocean and atmosphere remains the same, but the concentration of seed in the atmosphere will go down and in the ocean will increase.”
You seem to be missing the point. CO2 is no in ‘equilibrium’ it is in steady state. These are two different things. We know we are dealing with a steady state from the ‘sawtooth’ nature of the yearly CO2 cycle. We know from calculation, from 14CO2 measurements that the half-life is about a decade. If the CO2 in Earth were an equilibrium, life would be destroyed in short order. All mineralization of CO2, as oil, coal, carbonates would have removed all the atmospheric CO2, rather than almost all the CO2. Or, following a large volcanic period we would face 200-1000 years of 100,000 ppm CO2. Levels this high, so we are informed, would raise temperature to two zillion degrees and kill off all the animals.
The atmosphere is BIOTIC, if you want to classical physical chemistry analysis on it you will fail. Treat it as biological and its easy. Just do steady state analysis.
I now have high speed internet so the AIMS animated graphs work for me. I had no idea they had data all the way to July 2008. The animation looks a lot like ocean SST animations. Which we know have decadal oscillations. From a scientific view, many of the variables related to atmospheric CO2 have oscillations that are quite long. CO2 itself has seasonal and diurnal oscillations. It would be reasonable to ask if CO2 has a longer oscillation as well. But there is not enough years of data on a truly global scale (with all the sources AND sinks accounted for) for us to see if there are such long term natural (like 20 to 50 years) oscillations. It seems scientifically juvenile to jump to conclusions so early in the data collections process.
Using the Mauna Loa site to trumpet impending doom just doesn’t pass the scientific smell test. It would be just as foolish for me to state that Wallowa County is a good place to measure snow levels and then say it is representative of the entire planet. We now know that snow levels vary a great deal across the globe, and has short term and long term local and regional oscillations. I know this is a laughable example but logically, is it any different that the case with CO2?
Good science that leads to lasting, repeatable, testable and robust theoretical understanding of phenomena takes time. It seems that today’s scientists want to rush the process to what, be first in line for the big credits? Hope not. Take a lesson from playwrights. Your efforts have to pass the smell test at the local theater before they can hit the big city lights. If you don’t go through this process, your theory could end up on the one hit wonder late night talk show circuit.
doc martyn
yes exactly as I think. the 12 years for c14 co2 is time for ultimate removal rate, not a co2 half residency time.
We have no idea how much co2 is put into the atmosphere on a yearly basis, nor how much is removed.
but there are some natural experiments which are quite helpful. 4 times in the past 40 years, we have MLO data , and thus rates and we have carbon emission rate data. In these 4 periods, emission rate decreased. Each time MLO ppm increased but its rate increased, decreased or stayed the same. Each event happened at different MLO ppm. And different emissions rates being constant or decreasing.
In comparisons to various timeperiods, one would expect that as emissions rates doubled, for the time periods in which their rate of increase was steady and unchanging, that MLO would increase as the net of mans rate and the biosphere off rate. If mans rate were higher than a previous time periods emissions rate, and if the biosphere rate was the same, then MLO should increase more rapidly during these later time periods.
but it doesnt happen. the change in MLO during times of steady carbon rates is independent of the carbon rate and the MLO c02. This means that MLO is not dependent on the emissions rate
Steve Berry (14:28:01) :
“The other possibility is the global economic crisis. This has led to lowered consumption of fossil fuels, […]”
Sorry, but this is nonsense. Firstly, the amount of miles driven doesn’t appear to have been reduced during the economic crunch – there is no evidence to support this at all. Secondly, the addition of CO2 from vehicles is actually small. […]
REPLY: I was going by a report from the American Automobile Association in late 2008, citing reduced miles. I’ll see if I can find it again. But your point about contributions is well taken. – Anthony
You can actually short cut all of this and just go straight to global oil production figures. These are widely available. Bloomberg (and others) report oil inventory changes for the U.S. every Wednesday (about 10am PST) from Fed data. At that moment millions of dollars are won/lost/bet by traders. Oil production and inventory are a couple of the most widely followed data in the trading world.
Last I looked, OPEC had voted a couple of million bbl/day reduction from the 88 (or so) being produced world wide, but were having trouble meeting their goal. As a first approximation, total oil can be down between 1/88th and 1/44th from peak to now. Not exactly big. Oil consumption is very inelastic. That is why the price rises to $148/bbl when demand is high then drops to $38/bbl with a minor drop in demand. It’s not manipulation, it’s a very inelastic supply & demand curves.
From: http://www.eia.doe.gov/steo
Consumption. World oil consumption continues to be revised downward in response to the global economic downturn. Global consumption is estimated to have been largely unchanged in 2008 and is projected to fall by 800,000 barrels per day (bbl/d) in 2009.
A better place to go fishing for an impact would be the coal used in steel production and to power factories. This is directly tied to the economic cycle and the quantities are very large.
Aluminum (energy intensive) demand has fallen off a cliff (so AA stock is way down), and steel for autos, well, can you say 44% off? (X on the bottom, though maybe life soon?) The coal stocks (BTU, PCX, ACI, MEE, YZC, and even TCK a mixed metal /coal company) are all in the same place. Whacked down hard, looking like they will be working their way up over the next couple of years.
From the same report above: http://www.eia.doe.gov/steo
Coal
Consumption. The projected decline in electricity consumption, combined with projected increases from other generation sources (nuclear, petroleum, and wind) will lead to a 0.7-percent decline in electric-power-sector coal consumption, which accounts for more than 90 percent of total coal consumption. An expected increase in electricity consumption in 2010 of 1.5 percent will lead to a 1.9-percent increase in electric-power-sector coal consumption. Consumption growth in the coke plant sector is estimated to have been flat in 2008 but is expected to fall by 8.2 percent in 2009 and by 5 percent in 2010 due to the economic slowdown. Retail and other industrial sector coal consumption is expected to decline by 9.0 percent in 2009 but increase by 0.7 percent in 2010 as economic conditions improve.
Realize that due to contract issues, if I shut down my coking plant 6 months ago, I may well have had to take delivery on 6 months of coal and pile it on my lot (still ‘consumed’ as far as production figures are concerned) and that is why, IMHO, they show 2008 will be ‘flat’ but in 2009 ‘consumption’ will drop by 8.2%. Even if a recovery begins, I’ll be using that pile I was forced to store before I buy more. Expect the “consumption” figure changes to lag the actual burning of the coal changes by about 3 months to 1 year depending on where you are in the business cycle.