Thanks John, and yes, rather a lot of work...
That said, we kept to the largest, "most respected" (difficult terminology to use that when mentioning GISS, and MLO)
widely acknowledged, widely used, and publicly available for free data sets, as much as possible.
Within the limits of what the piece / pdf was to cover it could not go into MLO in much detail,
or the constant alterations to the GISS data set (I'm not going to give any of that obvious and
always in the same direction manipulations of the data set credence by calling them "corrections").
Or should that be GISS data sets by now......
However I did put a bit in the excel sheet regarding MLO.
I'll repeat (for the first time outside of the excel sheet) it here
as I think it is useful when considering Hypothesis 1, and also to some degree Hypothesis 2.
Contained in the freely available excel sheet is..(I'll have to update and replace some of
the contained links below, but the general gist is still correct in my opinion.)
Derek's personnal opinion of the Mauna Loa Observatory record of atmospheric CO2 purportedly global measurements.
Please feel free to copy and paste to forums, blogs, etc for further discussion.
We do NOT measure (global) CO2 levels.
How many elephants are in the room. ?
Mauna Loa Observatory has been measuring global CO2 levels for 50 years now, in January 2009.
How good a record has it produced. ?
In the first version I looked at this official page
of the method employed to measure the beneficial to plant life trace gas that is CO2.
How does the CO2 analyzer work? Air is slowly pumped through a small cylindrical cell with flat windows on both ends.
Infrared light is transmitted through one window, through the cell, through the second window, and is measured by
a detector that is sensitive to infrared radiation. In the atmosphere carbon dioxide absorbs infrared radiation,
contributing to warming of the earth surface. Also in the cell CO2 absorbs infrared light.
More CO2 in the cell causes more absorption, leaving less light to hit the detector.
We turn the detector signal, which is registered in volts, into a measure of the amount of
CO2 in the cell through extensive and automated (always ongoing) calibration procedures.
What motivates "MLO et al" in this monitoring program appears to be explained by Dr Tans
in one of his replies to Anthony Watts on this thread at WUWT
excerpt of one of Dr. Tans email replies to Anthony Watts ,
" We are very much aware that in a time
when carbon dioxide emissions will cost a lot of money, there has to be
an objective and fully credible way to quantify emissions. Without
that, carbon markets cannot function efficiently, and policies cannot be
measured relative to their objectives. We think that the atmosphere
itself can provide objective quantification. "
End of excerpt.
CO2 is called a trace gas because at a concentration of 385 parts per million this in percentage terms would be 0.0385%.
If the atmosphere were an elephant, then CO2 would be the head of a pimple, on the back of an elephant.
Put another way if you imagine the atmosphere as a whole as represented by 100 pennies then the amount of CO2 present would be,
the green coloured slice (somewhat exaggerated for clarity..) of one penny shown below out of one hundred pennies.
(I would of done a before the Industrial revolution CO2 level and present version for comparison,
but there is no difference visible as I have already exaggerated CO2 level depicted by more than the change over the last 150 years..)
The above linked to and quoted description of the method used by the Mauna Loa Observatory, (and up to 60 other stations in the official network),
which Ill refer to as MLO et al would appear at first reading to imply that an overall measurement is taken
of the greenhouse effect of the gases present in the dry air sample measured.
This is not the case.
The measurement is far more specific, namely the measurement is of the 4.255 um wavelength,
so my previous post is based on a very basic misunderstanding, and
is not valid in the respect of what is actually measured.
The first plot in the previous post is worth repeating here, namely,
The more accurate way to interpret this plot is that for a given wavelength
the spectral response times
the amount present of the particular gas
would added together give the sum of the gases contributing to
the measured effect at a particular wavelength.
In this thread
, Ferdinand Engelbeen kindly provided these levels for the GHGs
that might be expected to be roughly present in any well mixed air sample.
H2O: 1,000-10,000 ppmv
CO2: ~385 ppmv
CH4: ~2 ppmv
N2O: ~0.3 ppmv
O3: ~0.1 ppmv
CFC's: sum of several: less than 0.001 ppmv
In short, at the wavelength measured the amounts of other gases that overlap CO2 (spectrally)
are present in such small amounts and their responses at this wavelength are so small that they can be ignored.
I have to admit that excepting water, this appears to be the case, but even so the sum of the other gases,
(N2O and O3 I believe) that are contributing, and may (do) vary MUST be taken into account
if the purported measurement accuracy of MLO et al to 0.1 ppm is to be correct.
Given the way even my posts have been responded to, I would suspect constant corrections have been used,
but what they are is simply not known, as they have never been divulged by MLO et al.
There is however an elephant in the room.
If you consider the argument that it is the amount of a GHG present, how it varies, and
its spectral response at a specific wavelength that is the important factor overall then
something obvious lands on the discussion.
The water vapour elephant.
If you refer back to the pennies figure above you will notice spaces between the pennies. In a way the real atmosphere is like this,
the spaces according to temperature can be filled with more or less water vapour.
(Imagine the pennies moving away from each other as the temperature increases, and the reverse when cooled)
As Ferdinands figures above show water vapour content can be both considerable and varying.
In almost all studies it is now widely accepted that water vapour is THE most important GHG
(contributing between 85% to 95% of the so called greenhouse effect)
because of its strong and wide spectral response, and the considerable quantities involved.
From the descriptions so far it becomes apparent that MLO et al believe they have dealt with this elephant.
Again looking over the official pages it appears a constant correction / assumption of 4% is used.
The next line of justification usually employed is that the air sample is dried by the cold water trap.
To dry air completely a temperature of about minus 70 C is required.
This is why the cold water temperature traps temperature record is important.
Given the size of the (water) elephant even a slight change in the cold traps temperature could have
considerable and undesirable effects on the every ten seconds, accurate to 0.1 ppm measurements.
These QA / machine constants as far as Im aware have never been released, or independently verified.
This (water) elephant also does effect the air samples before they are even taken.
For instance in the cold of the early morning as dew forms the air cools as does the water vapour.
CO2 dissolves very readily in water, and the colder the water the more CO2 the water can hold.
In other words the CO2 / water solubility changes mean that the water vapour reduces the CO2 level in the air in the early morning.
It would seem reasonable that during a normal day CO2 levels would alter quite markedly due to
the changes in the airs temperature due to the solubility of CO2 in water relationship alone.
Rain would have an effect as well, lots of cooled water, high humidity.
Is there a temperature / pressure / humidity record for MLO.
Apparently not. ?
Are there any other elephants in the room. ?
The discarded outlier elephant.
The idea of discarding any measurements that over several hours vary by more than 0.25ppm is usually justified by MLO et al
because there is little or no variation measured at the South Pole this is plainly at best erroneous,
but MLO et al does this routinely.
(The South Pole being surrounded by a vast and cold ocean, goes from 24 hour daynight to 24 hour night,
no vegetation, one of the driest places on the planet, known to have a different climate to just about everywhere else,
etc, etc, etc, the South Pole is plainly different, not the same, not applicable.)
On Dr Glassmans comments
he summed this problem best with this statement.
Ferdinand reports that the data are reduced to produce smoother results.
He reports that investigators select data to reject volcanic, agricultural, and other local effects.
He says that they remove outliers before averaging, and that that is normal scientific and engineering practice.
Without more, these are all subjective steps. While they may be common practice,
they are at best incompletely reported or worse - unscientific.
Terrible mistakes have been made and actual frauds perpetrated by improper removal of outliers.
Interestingly massive mistakes have been made using South Pole measurements incorrectly in the past
in other climate science areas of environmental concern.
Ozone for example, the South Pole data / measurements of Ozone were used to show how man had created the ozone hole.
That idea is now all but discredited, as the processes / effects both natural and man made were not understood correctly.
As recent discoveries
The events of this sorry tale are well known, and are relevant to the MLO et al exclusion of outliers.
In the late 1940s and early 1950s some measurements were taken showing an Ozone hole that appeared to vary seasonally.
These measurements were later thrown out as being taken with unreliable / inaccurate instruments.
Some years later (when most people had forgotten about these earlier measurements)
a major company (DuPont I think)
had a patent that was about to run out on the refridgerant used at the time.
To protect the companys future it would be useful if the old fluid became illegal and the newer,
not as good, more expensive fluid the company had a brand new patent on came into widespread use.
The discovery of the man made Ozone hole was made in the late 1980s / early 1990s, using
exactly the same equipment as used in the late 1940s / early 1950s.
The reliability of the new measurements led to the Montreal Protecol.
The Montreal protocol.
Incredible, but true,
we have all had to buy more expensive, less efficient fridges, that make damned annoying pinging noises.
As if to rub salt into our wounded wallets the seasonal Ozone hole still appears to be
at about (within 1% of) record size.
Maybe it was the sun after all, at the South Pole, and the oceans plus the sun in the tropics.
Maybe, just maybe we did not need to buy new fridges and air conditioning units after all.
Back to MLO, are there any other elephants in the room. ?
The variation of what do we actually measure elephant.
Yes, spectrally we appear to measure CO2 but what and which variations that are not admitted already as discarded. ?
In the mid 1950s Keeling himself produced a paper that discarded the effect of the vegetation
to any great extent on the MLO measurements, more recently plots have been produced showing
a net decrease in the MLO vegetation changes rate confirming, or rather reaffirming Keelings original findings.
In short the vegetation appears to be on the wain (less active sun possibly)
yet the MLO official rates of change of CO2 have not changed.
Almost any reading of the official MLO sites refers to CO2 depleted (by vegetation) air samples.
There are other even more basic "what do we measure" questions.
Many measurements are flagged (discarded) by MLO because of the local volcanoe,
usually these are referred to as downslope wind directions, and nothing more.
In the case of downslope from the volcanoe winds this appears reasonable,
if it was correlated to volcanic activity or gas emissions.
Im not aware of such a tally, merely too high measurements from the wrong direction are discarded.
The volcanoe could be permanently emitting gas, but I doubt it is a constant, even so this may be reasonable.
Upslope winds appear to make up most of the voltage measurements that are kept for processing.
These winds are therefore used as our measurement of global CO2 levels.
Locally Keeling suggested vegetation was not much of an effect, but officially vegetation is given
as the reason for the seasonal variation as measured at MLO.
The changes from one season to the next showing the changes in global CO2 levels that man is supposed
to be altering so dramatically. Allegedly.
Let us not forget at this point the science is supposedly settled. ?
Is there anything else that could be effecting the measured CO2 level at MLO. ?
Mauna Loa is an island, it is in the Pacific ocean, and this island is near the equator.
Seasonally the trade winds shift North and South, as do the Hadley cells they are part of.
The tropics of Cancer and Capricorn show the seasonal movements of the Hadley cells North and South of the equator.
In a year the Hadley cells and their trade winds will move over Mauna Loa.
The official position is that CO2 is a well mixed gas, obviously not instantly but over many months globally speaking.
The usual answer given is that it takes about 18 months for CO2 emitted from any locality
to become evenly mixed globally.
Earlier in this post it was mentioned that CO2 solubility in water is heavily temperature dependent.
When warmed water de-gases CO2. This is seen every time you heat a pan of water,
the bubbles that form inside the pan before it boils, these bubbles are in a large part dissolved CO2
being de-gassed as the water as it is warmed.
Dr. Glassmans page The Acquittal of CO2
covers this relationship very well.
Untill recently even the PDO phases of the Pacific were not realized, and in most cases the discovery of
the PDO warm and cool phases of the Pacific ocean are not thought to have been discovered until 1997.
The trade winds that blow across Mauna Loa have blown across a part of the Pacific where the water is known
to be warming and so therefore de-gassing CO2. The longer the wind has blown across the ocean the more
CO2 that will have been de-gassed by the ocean, so the higher the level of CO2 measured.
Over the path of the trade winds it would be reasonable to expect an increasing level of CO2 as the wind neared the solar zenith and the doldrums
Below is a roughly drawn figure that shows this.
Obviously Mauna Loa being an island does not move, but the trade winds do move with the suns zenith,
between the tropics of Cancer and Capricorn. They move seasonally, regularly, and relatively predictably.
The point this figure is trying to illustrate is that MLO is on an island that the trade winds move over seasonally.
If you have a measuring station (on the above figure) at point A would you expect a lower CO2 reading than at point B on the diagram. Yes.
Is this the seasonal variation that MLO is measuring.
Is this another elephant. ?
The algorithm elephant
Quite simply, remember Manns 1998 and 2008 Hockey Stick temperature projections.
In the 1998 Hockey Stick there were supposedly 400 plus data sets. Then by accident someone found a copy of the algorithm Mann et al used.
It turned out that Mann had used 112 data sets only, AND one data set had been weighted by a factor of 392.
So, what appeared to be 400 plus data sets was in reality just one.
No one has seen the algorithms used by MLO et al.
We may have seen 2 days of 10 second voltage data of a 50 year record,
but algorithms, no, not at all, not one.
Constants / corrections used, what, how, why to calculate the ppm figures released,
we simply DO NOT KNOW.
So, how many elephants are in the room,
4 and counting.