Guest essay by Clyde H. Spencer
Before getting into the details of the analysis, it must be stated that there are a number of issues with the available temperature data sets. Some critics have dismissed the historical ocean pH measurements because of the poor spatial sampling. However, the same criticism can be made about historical temperatures. There are still concerns about the influence of the Urban Heat Island effect on recent temperature readings. In addition, the temperature record is a moving target because there are on-going changes made to try to correct for the shortcomings of a system that wasn’t intended to be used to track long-term changes. Therefore, if you attempt to reproduce my results, you will probably find that the data have changed. The Berkeley Earth Surface Temperature (BEST) project was started because of distrust in the other existing data sets. I have chosen to use BEST data (08-Jan-2016) because it is readily available and some consider it to be superior to the government data sets. Any conclusions need to be taken with reservation.
Unfortunately, the most detailed atmospheric carbon dioxide (CO2) concentration data that are available cover only about the last 56 years. The Mauna Loa CO2 data are available from the Scripps Oceanographic website. A scatter-plot of BEST monthly temperatures versus logarithm base-2 (Log2) PPM (parts per million) CO2 concentration was prepared for the period of 1958 through 2015. (See Fig. 1, below) The point cloud for the monthly data is not particularly tight. Indeed, the least-squares fit regression-lines only show an R2 value a little above 0.5 for both the average global high and low temperatures. The classic interpretation of the R2 value is that it represents the amount of variance in the dependent variable (temperature) that can be explained by the independent variable (CO2 concentration). That is, in this situation, only a little more than half of the rise in temperature can be related to the increase of CO2 in the last half-century. Of that CO2 increase, the combustion of fossil fuels probably only accounts for about 75%. (See Spencer, 2015).
Using raw PPM rather than the LOG2 of the PPM CO2 gives slightly higher correlation coefficient with the monthly high-temperatures. (0.533 v. 0.528). Whereas, the opposite is true for low temperatures (0.548 v. 0.550). Thus, it seems that the average low-temperatures behave slightly better with respect to the theory that CO2 is impeding radiative cooling.
Using a 2nd-order fit instead of a linear fit increases the correlations slightly. However, the trend-line curves upward with the high temperatures for the most recent data, but downwards for the low temperatures. The apparent opposite results, suggest that something is affecting the high and low temperatures differently. I have pointed out previously, (Spencer, 2015) that the difference between the high and low temperatures has been increasing in recent years and seems to be the result of the high temperatures increasing more rapidly than the low temperatures. That is the opposite of what one would expect if CO2 were the main driver.
Fig. 1 Monthly average temperature versus Mauna Loa CO2 concentration
Substituting the presumed concentration of CO2 (277 PPM) for the pre-industrial era into the linear regression equations yields high and low temperatures estimates of 13.537 and 2.207 degrees C, respectively. That gives an average of 7.872°C, compared to an average temperature of 8.367 for 1771 from the BEST database. It is a surprisingly good agreement for a linear prediction based on temperatures that generally appear to increase in a saw-tooth pattern.
While the correlation of temperature with CO2 looks compelling, what if it is because the CO2 is coincidentally highly correlated with something else such as natural, in-phase long-term temperature trends or it is simply a proxy for the totality of anthropogenic influences? Remember that correlation does not mean causation!
However, using annual Law Dome CO2 data from 1759 through 2015, the relationship is less well-behaved. (See Fig. 2, below) The correlation coefficient (0.5638) for average temperature versus CO2 is, approximately, what is observed for the Mauna Loa monthly CO2 data. Non-linear behavior is particularly striking for Log2 CO2 concentrations less than 8.35 (≈pre-1972). There also seems to be particularly serious problems with the estimate of CO2 concentrations (284 PPM) or temperature around 1805 to 1840. Note that the slope of the regression line is similar to those in Figure 1
Fig. 2 Average annual BEST global land surface-temperature versus Law Dome ice-core CO2
It is generally thought that pre-industrial CO2 levels were relatively constant, only showing very slow increases. However, at low levels of CO2, the temperatures were varying in a manner not expected from the theoretical model. Some of the low temperatures associated with the downward spike in temperatures with a CO2 log2 concentration of 8.15 may be the result of the eruption of Mt. Tambora (1815), but not all. The temperatures apparently started to decline a decade before the eruption, and remained low longer than the typical two or three years after a major eruption. The decline in temperatures resulting from the 1883 eruption of Krakatoa is barely discernible at a concentration of 8.18. There are also high temperatures paired with low CO2 concentrations.
Assuming that the relationship between CO2 concentrations and temperature is as shown in Figure 1, then the historical temperature data are not trustworthy. If the temperature data, which are 12-month averages, are correct, then there appears to be a serious problem with the assumed control of CO2 over temperatures! Alternatively, the CO2 concentration would have to have been varying considerably at this time to explain the different temperatures. This goes to the heart of my opening statement about the veracity of the historical temperature data and the ability to say anything about temperature increases for anything other than the modern record. Although it is not highly probable, in my judgment, one has to at least entertain the possibility that the modern rise in temperature along with CO2 is a coincidence.
A plot of estimated atmospheric CO2 concentration versus population for the period of 1958 to the present day shows that the rate of growth of CO2 is greater than the population rate. A 2nd-order least squares fit gives an R2 value of 0.999. (See Figure 3 below.)
Fig. 3 Annual Mauna Loa CO2 concentration versus world population
The correlation of historical, global CO2-increases, with population increase, is so high that one must entertain the possibility that the CO2 is a proxy for the totality of anthropogenic effects when used to predict temperatures.
One such effect is anthropogenic water vapor. Combustion produces water from all hydrocarbons, along with CO2. Water used in steel rolling mills, and many other industrial applications, evaporates under conditions it would not have done so were it not for Man. Similarly, water used to cool nuclear reactors and other power plants is released into the atmosphere; it initially condenses into visible water droplets, and then evaporates, increasing the relative humidity. Reservoirs in arid regions provide water vapor both from the reservoir surface and the fields the impounded water irrigates; the water vapor would not have been present before the dams were built. Lastly, massive depletion of underground aquifers, largely for agricultural irrigation, has brought water vapor into contact with the atmosphere during the growing season in arid and semi-arid regions to change the balance of the relative humidity. The essential point here is that ‘Greenhouse Gas’ theory predicts that increasing CO2 will warm the atmosphere slightly and cause additional evaporation of water, which amplifies the CO2 warming. Man is providing additional areas from which water can evaporate.
Another anthropogenic effect is Urban Heat Island contamination of the temperature records as the cities have encroached on what were formerly rural areas. The BEST project claims to have disproved that hypothesis, but it is my opinion that they didn’t search far enough outside the city limits, nor in the right direction. Quattrochi et al. (Project Atlanta, 1999) have demonstrated that the heat and pollution from central Atlanta (GA) influences the weather for miles downwind from the city. Additionally, Watts (2015) has demonstrated a woeful lack of adherence to standards in the siting of many temperature-recording stations. I don’t think this is “settled science.”
A plot of BEST global average land-temperatures versus world population (Figure 4, below) produces what appears to be a near-linear trend with an R2 value (0.782). However, fitting a 2nd-order polynomial produces a trend line with an R2 value of 0.811.
Fig. 4 Average annual global land-temperatures vs. world population
This suggests that all anthropogenic influences may account for as much as 81% of the variance in the land temperatures. I should note that if one plots 12-month smoothed BEST temperature data in Figure 1, instead of monthly temperatures, a linear correlation of similar magnitude is obtained. Therefore, the nominal R2 value of about 0.5 is likely an upper bound on the CO2 impact alone.
Atmospheric CO2 is characterized as a “well-mixed gas.” However, the NASA OCO-2 satellite shows a range of about 4% throughout the world, integrated over a 1-month period (see Fig. 5, below). That is approximately 10% of the claimed total increase in CO2 during the Industrial Era. Notably, there is no obvious evidence for the Northern Hemisphere industrial emissions. The preponderance of high values is over the southern oceans, which might be the result of out-gassing. The Amazon basin also shows elevated CO2; it is unclear whether that is a result of human burning activities or normal decay of vegetation. One needs to ask why OCO-2 isn’t confirming the presumed Northern Hemisphere anthropogenic CO2 when it is blamed for the historic temperature increases, and why there is a larger increase in high-latitude temperatures where CO2 has the lowest concentrations!
Fig. 5 CO2 concentrations from the OCO-2 satellite, July 2015 (NASA/GES DISC)
In summary, approximately 81% of the warming in the last century may have resulted from all anthropogenic influences, as suggested by figure 4. This includes water vapor, CO2, methane, nitrous oxide, and land use changes to the albedo and thermal mass. CO2 may account for as much as 52% to 56% of the contribution from anthropogenic drivers (See Figs. 1 & 2). Fossil fuel-CO2 represents less than 75% of anthropogenic CO2. If we were successful in completely phasing out fossil fuels over the next 100 years, we would have a reduction of 50% in average CO2 emissions. If the Earth is warming at a nominal rate of 1°C per 100 years from all influences, then we can hope, at best, for a reduction in temperature increase of 20% (0.54×0.75×0.50) or 0.20°C. That is to say, if the world were to phase out fossil fuels in the next 100 years the warming would be 0.80 degrees instead of 1.00°C! Unfortunately, eliminating fossil fuel use will probably not be successful in significantly reducing future temperature increases, even if it can be accomplished.
The relationship between atmospheric carbon dioxide concentrations over land and ocean, show most CO2 source increases are over the oceans. Despite all the CO2 humans are suppose to be releasing there is generally a lot less CO2 emissions over these areas compared to out-gassing CO2 over the warmest oceans, covering the Tropics and sub-Tropics.
The inconvenient relationship between CO2 and temperature over the last few million years is that when high CO2 levels are reached a Ice age results soon after. Therefore CO2 is a negative feedback and in fact the planet is full of negative feedback’s that keep the planet more stable than a humans body, yet the mass difference between them are gigantic. It is remarkable how such a huge body can remain stable thanks to many numerous negative feedback processes.
The biggest problem with land temperature and CO2 is that since the 1940’s there has been more decades with rising CO2, but no rising temperatures indicating natural cycles are dominating.
“It is remarkable how such a huge body can remain stable thanks to many numerous negative feedback processes”
Matt – yes, remarkable, like all evolution 🙂 Biologists have a very big part to play in this research too
The OCO2 color coded map shown is for the month of July, when there are great seasonal natural sinks in the northern hemisphere. The squiggle in a graph of monthly readings from Mauna Loa shows that seasonal alternations between sourcing and sinking from nature, such as from northern forests, outweigh manmade contributions. However, the seasonal sources and sinks largely cancel each other out over a year. What would be good is a color coded map of CO2 over the world over a whole year.
“However, the seasonal sources and sinks largely cancel each other out over a year.”
Says who? There is no physical necessity for this claim.
And, if they do not cancel each other out to within a gnat’s eyelash, they are so large that the residual would easily swamp human inputs.
Except that the residual is only half human emissions and never higher than the emissions…:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em2.jpg
This is circular reasoning.
‘Says who? There is no physical necessity for this claim.’
Well if it is seasonal. Please don’t be so aggressive.
Blog:
The graph is not from a year, it is a July snapshot. During the summer NH grows wood and grasses.
“Well if it is seasonal.”
Big IF. What is a season? Is every year exactly the same as before? Is the Earth always in exactly the same state at a given time of the year?
No, it is not. There is significant variation, some very long term.
Here’s the OCO-2 map for April 2015:
Phil,
The OCO-2 map you provided is labeled Sept. 2014 – Sept 2015, not April 2015. Did you put up the wrong map?
Clyde – you have to look at the bottom right – it’s 3/27 to 4/11. so actually two weeks in the Spring.
So, yes, apparently NH CO2 is elevated in the Spring, which is curious, since you’d think that the newly growing plants would be sucking it all in. So, one wonders what the source is?
Industry is year round yet, over the year, NH concentration plunges to the global minimum. So, all that red is not from industry.
Bartemis,
Those who fly aerial surveys try to pick a time when when the snow is gone and the leaves have not come out completely. That tends to be around March through May, depending on the latitude. At the time of this map, it is generally warm enough in the NH to encourage aerobic and bacterial decomposition of leaf litter left over from the previous growing season. So, without optimal photosynthesis to capture the CO2, I would expect the CO2 levels to be high. The origin is biological.
Hardly surprising given that the Mauna Loa data shows the annual peak is in May, also not surprising that when Spencer picked August that it showed low because that’s when Mauna Loa shows a rapid drop towards the minimum in September. The annual minimum is usually about 2ppm above the previous year’s.
Phil,
It wasn’t so much “picking” a particular OCO-2 map. I went out to the OCO-2 website looking for recent usable maps. The one I included in my post was the only one I came upon. I picked that one because I didn’t want to re-use the one I had previously used. (see http://wattsupwiththat.com/2015/09/15/are-the-oceans-becoming-more-acidic/ )
Better yet you could have chosen the annual movie:
https://youtu.be/_UEZqyGU5RU
Presenting the July data and expressing surprise that the NH values are low is rather disingenuous when that’s exactly what you’d expect from the Mauna Loa data
“And, if they do not cancel each other out to within a gnat’s eyelash, they are so large that the residual would easily swamp human inputs.”
This is what I suspect as well. I have been thinking about the argument between you and Ferdinand for a long time. It’s been one of the most interesting and entertaining discussions I have encountered in climate science – I really have to thank you both.
The problem with the mass balance as I see it is that CO2 can be thought of flowing between two reservoirs; biota that absorbs CO2 and biota that releases it (ie sources and sinks). If the rate at which CO2 is released through metabolism of dead organic matter by primarily bacteria in the soil is temperature dependant, but is not balanced by flora that absorbs it through the growing season you will see a rise in CO2. This may appear as a trend over time, particularly if the is a general rise in minimum temps since biota releasing CO2 are completely temp dependant and do not require photosynthesis.
If, however, man both emits a new source of CO2, plus reduces the capacity for CO2 to be absorbed, say by deforestation, PLUS local warming from UHI effects, then perhaps he could be a significant contributor. But it’s hard to find all this sort of using quantified, or directly observed/measured which IMO the only way to settle it for sure.
agnostic2015,
CO2 uptake by plants exceeds release by decay, feed, food with ~1 GtC/year:
http://www.sciencemag.org/content/287/5462/2467.short
and
http://www.bowdoin.edu/~mbattle/papers_posters_and_talks/BenderGBC2005.pdf
CO2 uptake by the oceans exceeds release with ~3.5 GtC/year nowadays (~2.2 GtC/year in reference year 1995):
http://www.pmel.noaa.gov/pubs/outstand/feel2331/mean.shtml
Not only are these studies conducted with a strong confirmation bias, not only are they undertaken in specific sampled areas and therefore not likely to be representative, but again, uptake is partially driven by anthropogenic inputs, so this is the pseudo-mass balance fallacy in another guise.
Bart:
uptake is partially driven by anthropogenic inputs, so this is the pseudo-mass balance fallacy in another guise
In my informed opinion, based on all available evidence, the net sink rate is near fully driven by the accumulation of human emissions…
Linearity of the CO2 sink processes:
e-fold decay rate for the excess CO2 above steady state In 2012:
110 ppmv / 2.15 = 51.2 years or a half life time of 38 years.
The figures for 1988 (from Peter Dietze):
60 ppmv, 1.13 ppmv/year, 53 years, half life time 39 years
For 1959:
25 ppmv, 0.5 ppmv/year, 50 years, half life time 37 years
The e-fold decay rate of slightly over 50 years is too slow to remove human emissions in the same year as emitted, as can be seen in the graph above. Thus that accumulates in the atmosphere (as mass, not the original molecules). That means that besides a small contribution caused by temperature, the whole increase in the atmosphere over at least the past 55 years is from human emissions and thus also the main cause of the net sink rate…
Ferdinand – this is just constructing a narrative, an interpolation of data to adhere to a prescribed scenario. There is no physical reason or evidence that compels that things must be as you say.
You need to consider alternative explanations. If you put as much energy into that as you do in trying to find consistencies with your preferred outlook, you would quickly find that the realm of possibilities is large, and there is no general preference for the one you have latched onto.
Bart:
There is no physical reason or evidence that compels that things must be as you say.
Henry’s law says that for a given temperature there is a fixed ratio between CO2 in the atmosphere and in the ocean surface. If the pressure in the atmosphere is higher than that ratio, there will be a CO2 flux from the atmosphere into the ocean surface in ratio to the pCO2 difference between atmosphere and surface,
The net sink rate over the past 55 years shows a nice linear ratio with the pCO2 difference between the atmosphere and the “equilibrium” pCO2 for the momentary temperature.
Seems to me that this is based on a very good physical reason…
“Henry’s law says that for a given temperature there is a fixed ratio between CO2 in the atmosphere and in the ocean surface.”
No, it does not say that at all. Henry’s law states that:
Bart:
No, it does not say that at all. Henry’s law states that:
“…water takes up, of gas condensed by one, two, or more additional atmospheres, a quantity which, ordinarily compressed, would be equal to twice, thrice, &c. the volume absorbed under the common pressure of the atmosphere.”
That is the same as saying that if there is a pressure difference between atmosphere and ocean surface, there will be a flux between the two in the direction of the lowest pressure. That flux will be in ratio to the pressure difference… No matter if that is static or dynamic, no matter if that varies by the minute and never is in steady state.
The oceans and Earth’s atmosphere are never in steady state. In particular, there is a huge input of new CO2 into the surface system every second from upwelling waters, and it has to be dispensed with, or it will accumulate.
There is a huge continuous input of new CO2 at the upwelling zones into the surface system that is simply following the water flow towards the sinks and is hardly dispensed in the rest of the water surface system. Thus never accumulates in these waters, but is released into the atmosphere in ratio to the pCO2 difference between water and atmosphere.
If the temperature increases, more is released into the atmosphere and less remains in the water phase in the stream from upwelling to downwelling. If the pressure in the atmosphere gets high enough at ~16 ppmv/°C, the CO2 in and out fluxes between the water flux and the atmosphere and thus between upwelling and downwelling are restored to what they were before the temperature increase…
Donald,
So are you saying the the vegetation in the Northern Hemisphere is more than capable of ‘neutralizing’ all the anthropogenic emissions during the growing season? The first OCO-2 map released during the AGU meeting was for the months of October and November and looked rather similar. The only place I could see what might be anthropogenic CO2 was in an area of eastern China. I’m surprised that there hasn’t been more released on what OCO-2 has been finding.
Why warming is greatest in the far north where CO2 is the least? CO2 increased everywhere, including parts of the world where there is less at the time of measurement due to seasonal factors, or less yearround because the colder oceans sink more CO2.
The Arctic and near-Arctic warmed the most because the surface albedo feedback is greatest there.
Donald,
“…surface albedo feedback” I have some problems with that. The Arctic is nominally dark almost half the year, with very low sun altitudes during the time it is in sunlight. It is notoriously cloudy there. Lastly, the oft made claim that open “black” water absorbs the sunlight strongly ignores the fact that it looks dark because the reflection is specular. Fresnel’s equation demonstrates that reflectivity can reach 100% with glancing rays. One has to take all these variables into consideration to calculate how much energy Arctic waters can actually absorb.
You have to take polarization into account too, the p=polarization has substantially higher absorption than the s-polarization.
Phil,
The sunlight coming in is unpolarized so the total reflectivity is unaffected.
If that were the case there’d be no market for polarized glasses! Take a polarizing filter and look towards the ocean with sunlight glaring off it, rotate the filter and you’ll find an angle at which it disappears, the one polarization which is preferentially reflected is eliminated.
You misunderstood. I didn’t say that the reflected light wasn’t polarized. It is split up into two components whose sum determines the total reflectivity. Were that not so, there also would not be a market for polarized sun glasses because you wouldn’t see anything through them. (at least not the surface from which the partially polarized light is reflected. There is only one unique angle for which the light is completely polarized, the Brewster Angle. It is unusual for one to be in the correct position to perceive light at the special angle, and because the sun is in constant motion, it doesn’t last for long. But, this is all a non sequitur because the important point is that the reflectivity varies with the angle of incidence, which is always large at the poles and sometimes achieves a glancing angle with 100% reflectance.
To naturally assume that there is a link between both CO2 and Temperature rises is incorrect. Convenient but totally incorrect.
1. There is no agreed relationship, even the IPCC state this. It is garbage.
2. If someone in the atmospheric science industry disagrees with this comment they need to show proof.
3. The burden of proof is with those that propose the warming theory, not others to disprove it.
4. The use of the OC-O2 for July is misleading, it is one month and a full understanding of the annual global transport is required
5. The current warming is mainly in the mid to far north latitudes, and along the annual atmospheric transport corridors.
6. This is not CO2 related. There are other factors at play.
7. The carbon cycle as depicted is a myth
There are so many mistakes in this post. I don’t know where to begin.
I will start with a general overview of what is wrong with the approach and in a later comment if I get time I will address every mistake
First and foremost ANY exercise that attempts to regress the temperature against explanatory variables must take account of the actual THEORY behind global warming.
That Theory says this: The temperature of the SURFACE is a result of EXTERNAL Forcings and internal variability. It is a function of ALL forcings– not just C02. C02 happens to be a large component but it is not the whole story. Over long periods internal variations will sum to zero, because energy cannot be created from nothing.
So here is how we fit the data
http://static.berkeleyearth.org/pdf/annual-with-forcing.pdf
This is a fit of C02 concentration AND volcanic forcing.
Here is the data: Now, for those of you who want to get a bit more detailed we actually treat C02 as a proxy for all human forcing . See the spread sheet below for the correlations between c02 and other forcing
http://static.berkeleyearth.org/xls/forcing-comparison.xlsx
In short, The first model we did had all forcings in it. It essentially matches a c02 only “fit” . You can do it either way.
Now of course the problem here is that we are Fitting only the LAND data.
And of course you can USE that regression to predict what the other 70% of the globe will show.
Since the ocean doesnt follow the land PRECISELY you will see some differences.
Then. Here is what Clyde needs to do to test his model.
Predict the temperatures of the Ocean using his model. Guess what the population over the ocean is?
Well DUH.
He could also do this to see that his regression is flawed.
1. Grid the population: You can use Hyde Population ( 5 arc minute data ) all the way back to 1750
what you will see is zero relationship between population and temperature.
2. For the US he could have used 1 arc minute data ( 1 sq km ) going back to the early 20th century
what you will see is ZERO relationship.
3. You could pick out all the stations with ZERO POPULATION and just look at those. Did that
No population effect.
“Over long periods internal variations will sum to zero, because energy cannot be created from nothing.”
How long? Days? Years? Decades? Centuries?
This is an absolutely incoherent statement without further qualification. There is a continuous stream of energy coming into the system from the Sun. All it has to do is accumulate. In order to reach a steady state, it has to dissipate at the same rate it is being introduced.
But, there is no reason that the rate of dissipation has to remain constant and, to the degree it does not, stored energy will either increase or decrease. The rate of dissipation is governed by storage mechanisms which have modal responses modulated by time dependent functions. The timelines of these modal responses stretch over many characteristic lengths, some of which are, indeed, centuries long.
Agreed, this is part of the “internal oscillation” fallacy. Simply calling it “internal” does not mean it has to be net zero. Neither does simply labelling it an “oscillation”. That is substituting trivial naming for any scientific logic.
For example El Nino / Nina is NOT a pendulum. It is two very different processes that act in opposite senses. That does not mean it is net zero.
La Nina controls the amount of solar energy input to the tropics. El Nino transfers OHC to the atmosphere, and ultimately to space ( it is ironically a cooling event terms of the Earth energy budget ) .
We have very little idea what CAUSES either of the processes or what triggers strong El Ninos like the present, so any suggestion the sum of these two processes is baseless.
Simply using the term “internal” to describe something offers no information that would justify the conclusion that it is net zero.
Thanks Mosh’ , a good start.
Let’s add that linear regression on two error laden variables not a legitimate use of the technique, despite wide applicaiton of similar techniques in climatology papers:
http://climategrog.files.wordpress.com/2013/11/ols_scatterplot_regression2.png
http://climategrog.wordpress.com/2014/03/08/on-inappropriate-use-of-ols/
If Clyde were to plot each part of figure 1 at full scale on each axis we would that there is massive diffuse cloud of data with little linear relationship. The OLS fitted line would almost certainly be wrong to they eye. Datasets with such high non linear relationships or “noise” will produce an erroneously low OLS fit.
There is nothing more than the most general similarity to a long term upward trend in both datasets. There is nothing that suggests anything less trivial which may be evidence of a causal relationship.
The first thing to do when quoting an R^2 value is to calculate what number would indicate a 95% probability of this being non random events based on the number of degrees of freedom ( data points ) used.
Since both CO2 and temperature are auto-regressive and essentially accumulative integrals, they will likely show some trivial similarity. It would be more convincing to examine dT/dt vs d/dt(logCO2) , that would show if there is anything more that trivial up or downward trend linking the tow datasets.
Steven,
I’m glad to see that you treat CO2 as a proxy for all anthropogenic forcings, which is the conclusion I made from the high correlation with population. You have missed an important point. I have not claimed that the population is directly responsible for the temperature changes. I have claimed that there appears to be a relationship between population and human activities that produce well-mixed gasses that affect land, air, and water. Because of the different heat capacity of water, land, and air, it is going to be more difficult to demonatrate a correlation between SST and land population. But, I would expect to find one. I used land data because I expect it to be more responsive to changes in forcings. You complaints seem like little more than sophistry.
“There are so many mistakes in this post. I don’t know where to begin.”
Me neither
“First and foremost ANY exercise that attempts to regress the temperature against explanatory variables must take account of the actual THEORY behind global warming.”
So I’ll just chime in on that. I see this over and over here. Temperature should correlate with CO2. Or with emissions. Ignoring what the theory actually says.
A kettle on a gas stove. It’s heated by the gas, right? So the temperature should vary as the gas setting. Well, say you are boiling the kettle and you try varying the gas. Does the temperature correlate with the varying gas setting? Very poorly. It might correlate with the cumulative gas burnt. That is different arithmetic.
GHG concentration creates a forcing, which is like the gas setting. It’s a flux. Temperature responds to cumulative flux. There is likewise no reason to expect correlation to flux itself.
OK Nick,
let’s take your kettle analogy and run with it. If you frequently turn the gas up and down and take temperature readings less frequently than you adjust the flame, you will see little correlation. However, if you go through a process of setting the knob low, taking a temperature reading, and then repeating the process with slight increments, you WILL see a correlation between knob settings and temperature. Now, you remarked you might see a better correlation between total gas consumed rather than the height of the flame. Indeed, If I were to plot Delta CO2 against temperature change I would expect to see a lot poorer correlation than using the integrated CO2 concentration. So, what’s the problem? My complaint basically is that the historical CO2 and temperature data are unreliable because they don’t follow the same predictive model that modern data does.
Clyde,
No, it’s not an issue of frequency of measurement. Suppose you sinusoidally (sin(t)) vary the heat flux (around a mean), and the well-mixed kettle responds with temperature equal to the integral, which is linear plus cos. The sine and cos will be basically uncorrelated, and the sin with the linear heating also not, except for endpoint effects, which could equally be positive or negative.
Nick,
Your sinusoidal knob is not a good analogy for the real world. You have to add a trend to make it reasonable. Then you will find a correlation between the trend and the “linear plus cosine.” If you de-trend the the two, you should find an excellent correlation between the input and output, just 90 degrees out of phase. But, what you observe WILL be influenced by the frequency of the sampling. You will only be able to detect changes that are 1/2 of the frequency of the sampling rate. In the real world, we have an opposite problem that there may well be periodicities that are much longer than our life times and are difficult to recognize.
“the well-mixed kettle responds with temperature equal to the integral,”
It varies with sin(w*t-atan(w*tau)), where tau is the thermal time constant, and w is radial frequency, which you have assumed to be 1. If the time constant is short, the response tends toward the instantaneous. If long, it tends to the cosine.
What time constants are we looking for? If heating of the oceans, that is exceedingly long, but if the atmosphere, that is fairly short.
Clyde,
“You have to add a trend to make it reasonable.”
Well, even if you have trend alone, you don’t get perfect correlation. If flux is t, and temperature response is proportional to the integral t^2, and continuously measured, the correlation is R^2=15/16=0.9375. But any variation from that will generally make it worse.
Nick,
“Temperature should correlate with CO2. Or with emissions. Ignoring what the theory actually says.”
I see skeptics make this mistake over and over again. It really is quite stunning.
Clyde. As Feynman noted the easiest person to fool is yourself.
You have fooled yourself into believing that you UNDERSTAND the theory you are trying to criticize.
Steven,
How do I know that you haven’t fooled yourself? Do I take your word for it?
Steven,
I finally had time to look at your spreadsheet. All I saw was a “CO2 proxy” and volcanic forcings. I didn’t see any attempt to partition any other antrhopogenic forcings, positive or negative. I’ve forwarded the thesis that anthropogenic water vapor is a significant forcing over land. After all, even the IPCC claims that CO2 only acts like a trigger and the increased temperature from it causes increased evaporation of water which does the heavy lifting, if you’ll pardon the pun. Basically, I’m saying that 56 years is a short time to conclude CO2 is driving temperature changes, and attempts to go back in time reveal problems with the historical temperature and/or CO2 estimates, based on the recent relationship between temperature and CO2.
“I’ve forwarded the thesis that anthropogenic water vapor is a significant forcing over land”
==========================================================================
FWIW, I would think it only potentially significant on the surface land record, not on the overall atmosphere over land. I do appreciated your approach.
P.S. Thanks for the Salby link.
This analysis seems to lack focus. This is a holistic physical problem. If a theory is incorrect it will generate paradoxes. The paradoxes all go away when the correct theories and there mechanisms are applied to explain the observations and the analysis results.
Summarize everything we know about this problem. Let’s get everyone on the same page which is to at least understand and remember the paradoxes. The paradoxes point to correct answer.
A single paradox is sufficient to invalidate a theory. When there are piles and piles of paradoxes that invalidate a theory the issue is not is or is not theory incorrect, but rather why the field in question is ignoring the paradoxes.
1) The no warming for 18 years paradox.
It is a fact that two independent satellite data sets which agree with hundreds of thousands of weather balloons show unequivocally that there has been no significant warming for more than 18 years at a time in which atmospheric CO2 is rising year by year. That is paradox. A paradox disproves a theory. The warming in the last 150 years was not caused by the increase in atmospheric CO2. There are dozens of other observations and analysis results that support that assertion.
2) The latitude warming paradox. If CO2 was the causing the warming – ignoring the water vapor/CO2 overlap white lie – the observed warming should be the same for all latitudes as atmospheric CO2 is more or less evenly distributed in the atmosphere. The observed warming is primarily high latitude and there is twice as much warming in the high latitude Northern hemisphere. This pattern of warming is a paradox for the CO2 forcing theory. This same pattern of warming has occurred again and again and again in the paleo record with correlating changes to cosmogenic isotopes. The changes in cosmogenic isotopes are caused by solar cycle changes.
http://arxiv.org/ftp/arxiv/papers/0809/0809.0581.pdf
3) The phase paradox. 7 out of 8 times planetary temperature rises and then atmospheric CO2 rises, as cause must lead effect the phase analysis result supports the assertion that the warming is not caused by the increase in atmospheric CO2.
Phase analysis of CO2 changes vs temperature vs anthropogenic CO2 emissions supports that assertions:
1) The recent temperature rise is not caused by increase in atmospheric CO2
2) The increase in atmospheric CO2 is not caused by anthropogenic CO2 emissions. Independent analysis of C12/C13 changes in the atmosphere supports the assertion that majority of the increase in atmospheric CO2 (no less than 75%) is caused by the increase ocean surface temperature and increased release of low C13 deep earth CH4.
Comment:
If the heat hiding in the ocean hypothesis is correct then there is sustained mixing of surface ocean water with deep ocean water. The silly Bern model of CO2 sinks and sources which is a white lie model created to create an anthropogenic CO2 issues, assumes there very, very, limited mixing of the deep ocean and surface ocean water. As there is 32 times more dissolved CO2 in the ocean than in the atmosphere, if a portion of the deep ocean is replaced with surface ocean water (this must occur if there is mixing) then there is a vast sink and source of CO2 which works to resist surface forcing changes in CO2 due to volcanic activity or lack of volcanic activity and due to anthropogenic CO2 emissions.
http://www.tech-know-group.com/papers/Carbon_dioxide_Humlum_et_al.pdf
4) It is pathetic in your face basic errors the 1 dimensional no feedback calculations (which were do more than 20 years ago) are ignored. If the white lie incorrect assumptions are corrected the surface warming for a doubling of atmospheric CO2 is reduced by a factor of 16 from 1.2C to 0.075C. This result explains why there are periods of millions of years in the paleo record when atmospheric CO2 is high and the planet is cold and vice versa.
Scientific Explanation why there is almost no warming for a doubling of atmospheric.
The One Dimension, No Feedback Forcing Calculation’s Deliberate Incorrect (White Lies/Fibs)
Comment: The so called 1 dimensional no feedback calculation shows ‘surface’ warming of 3.7 watts/m^2 or 1.2C. The following shows there is peer reviewed analysis that indicates that 1 dimensional no feedback calculation is too high by roughly a factor 16 due to white lie incorrect assumptions. The so called general circulation models (3 dimensional models) have more than a 100 subject parameters that can be adjusted to give any answer possible. The 1 dimensional calculation is however the basis for the entire charade.
P.S. Reducing the surface warming for a doubling of atmospheric CO2 by a factor of 16 from 1.2C to 0.075C. Now as half of the warming should have all ready occurred, this means only 0.035C or less than 5% of the 0.8C warming in the last 150 years can be attributed to the increase in atmospheric CO2. The explanation of the remaining warming is due to solar cycle changes.
Assumptions
A) Lapse Rate Fib
The so called 1 dimensional, no feedback, forcing calculations for a doubling of atmospheric CO2 ignored the fact that the lapse rate decreases when atmospheric CO2 increases which reduces the forcing by a factor of four. The change in the lapse rate is due to the fact that hot air rises causing cold air to fall causing the phenomena which is called convection.
B) Water Vapor Fib
The 1 dimensional no feedback calculation CO2 forcing warming for a doubling of atmospheric CO2 was done with no water vapor in the atmosphere. As the planet is 70% covered with water there is a great deal of water vapor in the atmosphere. As the absorbtion spectrum of water and CO2 overlap, water vapor in the atmosphere reduces temperature increase due to the doubling of atmospheric CO2 also by a factor of four.
Due to Fib A and Fib B, the warming due to doubling of atmospheric CO2, no feedbacks is 16 times smaller 0.075C rather than 1.2C which is so small the without feedback warming is the same as the with feedbacks warming.
http://hockeyschtick.blogspot.ca/2015/07/collapse-of-agw-theory-of-ipcc-most.html
https://drive.google.com/file/d/0B74u5vgGLaWoOEJhcUZBNzFBd3M/view?pli=1
Transcript of a portion of Weart’s interview with Hansen.
In addition to ignoring the fact that ‘greenhouse’ gases increase convection which reduces surface warming by a factor of 4, the without ‘feedbacks’ calculation also ignored the fact the absorption spectrum of water vapor and CO2 overlap. As the earth is 70% covered with water there is a great deal of water vapor in the lower atmosphere particularly in the tropics.
Redoing the double atmospheric CO2 level, no feedback calculation with a atmospheric model that takes into account the amount of water vapor in the atmosphere and the radiation effects of water/CO2 absorption overlap reduces the surface forcing for a doubling of atmospheric CO2 from 3.7 watts/meter^2 to 1.1 watts/meter^2 ( also reduces surface for a doubling of CO2 by a factor of four). The 1.1 watts/meter^2 increase in forcing will result in surface warming of ball park 0.1C to 0.2C which is so small, the no feedback case is the same as with feedback case.
Check out figure 2 in this 1986 published paper that notes the 1 dimensional calculations were done for a dry atmosphere which is physically incorrect. The 1986 paper notes the surface forcing is reduced by a factor of four if it is redone with the estimated water vapor in the atmosphere.
http://journals.ametsoc.org/doi/pdf/10.1175/1520-0469%281982%29039%3C2923%3ARHDTIC%3E2.0.CO%3B2
Perhaps some scientists lack an appropriate skill set to explain climate change. Many apparently even fail to grasp that a forcing must be time-integrated to produce temperature change (this article is one example). If CO2 is a forcing on temperature, its effect on temperature must be in accordance with the time-integral of CO2 (or the time-integral of a math function thereof).
CO2 has a very tiny effect on average global temperature (AGT). Ignoring CO2 completely but including the factors that matter results in a 97% match since before 1900. This is in spite of the fact that some agencies have changed the data to corroborate a Global Warming agenda.
The factors that matter are identified and the tiny influence of CO2 is quantified at http://globalclimatedrivers.blogspot.com
The apparent correlation between CO2 and temperature is entirely superficial and spurious. The actual direct correlation is between temperature and the rate of change of CO2.
http://woodfortrees.org/graph/esrl-co2/derivative/mean:12/from:1979/plot/rss/offset:0.6/scale:0.22
The arrow of causation is from temperature to CO2, and not the other way around. Human inputs have very little impact.
Bartemis says: “The actual direct correlation is……”
..
Then posts a graph showing the correlation.
…
The Bartemis says: “The arrow of causation is from……..”
…
Bartemis makes the common mistake of thinking correlation proves causation.
This is not a superficial correlation of low information, low polynomial order. It is a correlation that exists in every nook and cranny. The odds of being able to do that across the entire frequency spread without there being a cause and effect relationship are vanishingly small.
The arrow of causation is necessarily from temperature to CO2 because this is a derivative relationship, and it would be absurd to posit that temperature responds to the rate of change of CO2, and not its absolute level.
Again Bartemis says: “It is a correlation that exists…”
…
Sorry Bartemis, correlation does not prove causation.
..
Even you know that.
Christopher Keating,
After years of discussion wit Bart, I know that he is immune for any argument or observation that doesn’t fit his theory…
His correlation between temperature and CO2 rate of change is entirely from the influence of temperature on (tropical) vegetation, but vegetation is a net sink for CO2 over periods longer than 1-3 years. Thus there is no connection at all between the process that causes the variability and what causes the trend. As human emissions show twice the slope of the increase in the atmosphere, and fit all observations, these are the obvious cause of the increase in the atmosphere, not temperature. One can fit the same graph as Bart did as well as for variability and slope by combining both influences:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/rss_co2_emiss_deriv.jpg
Background at:
http://www.ferdinand-engelbeen.be/klimaat/co2_variability.html#The_real_world
This is no accident:
http://i1136.photobucket.com/albums/n488/Bartemis/temp-CO2-long.jpg_zpsszsfkb5h.png
Yes Bartemis, it is no accident that the noise in the CO2 signal is correlated to temperature. Now, besides your correlation graph, what other evidence do you have?
It’s not just “the noise”. It’s the whole kit and kaboodle. Long term, short term, every term.
Bartemis, unless you have other evidence besides a single graphical display of correlation, it doesn’t matter if it is long term, short term, or ANY term.
OK, Christopher. Sorry, you are wrong. But, I have see no point in continuing this silly conversation.
Bartemis, you posted: “Sorry, you are wrong”
…
I’m wrong to ask you for additional evidence?
..
Why is asking you for something besides a graph of correlation “wrong????”
..
Why is asking you for supporting evidence “silly????”
Bart,
Disregard Keating. He has no credibility, as can be seen here.
David, it doesn’t matter what you say, poor little Bart got caught confusing correlation with causation.
..
..
..
@ur momisugly C. Keating: You are either very dull or are intentionally mischaracterizing what Bartemis said today at 10:20am:
Bartemis
(emphasis mine)
********************************************************
********************************************************
@ur momisuglyFerdinand Englebeen: re: After years of discussion with Bart, I know that he is immune from any argument or observation that doesn’t fit his theory…
And here YOU are, SEVEN years later, still with your one-way radio blasting us with the SAME ideas that have been refuted by MANY (not just Bartemis) able scientists on this blog. Does it not give you pause when you see that MOST (if not all) of your most staunch supporters are rabid AGWers (I know that you are not one of those)?
A voice from SEVEN years ago:
( http://wattsupwiththat.com/2009/02/24/watching-co2-for-the-last-12-years-by-hemisphere/#comment-98306 )
Law of Nature
(emphasis mine)
Mr. Englebeen,
You and I both know that Bartemis has OFTEN given you ad arguendo concessions in his replies to you. He regularly admits when he disagrees with your conclusion, but cannot prove his absolutely.
Such a remark is beneath you. You are better than that! I know. You have been on WUWT many times. It is late where you live, so, if you don’t reply, I will understand.
Sincerely, SIN-CERE-LY, indeed, (smile)
Janice
Janice…..you say: “You are either very dull or are intentionally mischaracterizing ”
…
No mischaracterizing at all. Bartemis said, “The actual direct correlation is between temperature and the rate of change of CO2.” then he posted a plot of the correlation.
..
..
CORRELATION
…
Now Janice, you know full well that correlation does not prove causation, right?
..
Tell me how a direct quote from Bart is “mischaracterizing” what he said?
C.K.:
When
assuming causation is logical. Bartemis did not assert that he had proven causation, only that it APPEARS highly likely to exist and that the correlation is so strong as to make a prima facie case for his point.
Thus, the burden of proof has now shifted to
YOU, C.K.,
to meet Bartemis’ argument with evidence.
Bart,
You appear to be unaware of the fact that HadCRUT is a work of science fiction.
Janice, please stop trying to prove causation using correlation.
..
The problem Bart has is that he has no additional evidence beyond the correlation. You know full well that not only is there a strong correlation of CO2 with recent global temperature, backed up by the physics of the green house effect. Bart needs to show evidence of ocean out gassing to supplement his claim of causation. He has not provided any such evidence. His problem is he’s hanging his hat on a single correlation found between the noise in a CO2 signal and global temperature anomaly. When and if he can provide physical evidence that this causal chain is in fact occurring, he might be listened to instead of being ignored for his arm waving
I have no evidence that gravity causes objects to fall beyond the fact that they consistenly do.
There is no strong correlation between atmospheric CO2 concentration and temperature beyond the fact that both were going up for much of the 20th century. Having them move in the same direction is not a strong correlation. It is a 50/50 coin toss.
Thank you Bartemis for admiring that your suposed “relationship” is nothing more than correlation.
…
When you have evidence of causation, get back to me.
..
PS….when you can show General Relativity (i.e. gravity) is only correlation please publish.
Correlation is a necessary but not sufficient condition to assert causation.
Also needed is a falsifiable theory to test and what is commonly called “face validity”.
One factor providing face validity is that the effect is preceded by the theoretical cause.
I don’t see any assertion that correlation proves causation here by Bartemis. I also disagree that the mistake of thinking correlation proves causation is in any way common in the posts on this site. However, could it be said the correlation is between the rate of change of temperature and the rate of change of CO2?
______________
This article by Mr Spencer seems to assert that a CO2 increase results in a temperature increase on the basis of an R2 of 0.5 using OLS regression. While OLS is robust with regard to violations of its underlying assumptions, 400,000 years of ice core data indicates that CO2 levels lag temperature changes; thus the model is misspecified, and the R2 is meaningless.
In other words, the causal relationship between CO2 change and global warming is spurious.
If this is a misunderstanding of Mr Spencer’s article, please accept my apology.
BTW, a comment was made by Mr Steve Case about the MWP being about 800 years ago, which agrees with the average lag time between global temperature increase (as provided by ice cores) and the rise of CO2 levels.
I’m not sure about the point of his comment, but CO2 levels are indeed rising.
Thomas,
I’m aware of the claim that ice core data supports the claim that CO2 levels followed the increase in temperature. However, from my reading it appears to be a contentious claim. I didn’t want to wade into that based on a single data point. I’m also aware that as the firn collapses to form ice there is considerable compression and the entrained air is relatively free migrate within the open pores of the firn. Thus, I’m not convinced that the claim is reliable. Also, as a general rule of thumb, one can expect that the resolution of time decreases the farther one goes back in time. So, I didn’t want to take a strong position on that.
“…when you can show General Relativity (i.e. gravity) is only correlation please publish.
Pretty much. It entirely rests on the assumption that the Ricci tensor is zero in the vicinity of a massive, neutrally charged, object, and correlates things like the gravitational red shift, the precession of planetary orbits, and the decay of the orbits of neutron stars to that assumption.
But, you don’t really know that, do you? You don’t even know what the Ricci tensor is. You simply think that talking about Relativity will make people think you are smart. I, on the other hand, can derive it all from scratch, and use it in my job as an everyday thing. So, how about you just run along, Junior. Nobody cares about your pretentious and hollow pronouncements.
Bottom line: CK, the burden of proof is on you and all the other AGWers to provide even ONE quantitative measurement that makes the conjecture about human CO2 emissions rise even to the level of a valid hypothesis. So far, not one piece of data proves causation. The climate models’ projections are anti-proof, they failed so badly.
And even CORRELATION is against you: CO2 UP. WARMING STOPPED.
******************************
And regardless of Bartemis’ ignoring your mischaracterizing him, I and many others can see that that is exactly what you did. And it is evidence, out of your own mouth, that you are dishonest. Thus, you have damaged your credibility here to the point that nothing you say can be taken seriously.
Janice, you say, “Bottom line: CK, the burden of proof is on you :
..
Nope….I have no burden of proof. I merely pointed out the pathetic attempt by Bartemis to show his “correlation” is causative. Janice, really, Bartemis has nothing other than his little graph. He has no other evidence for HIS CONJECTURE. I don’t have any burden, all I have done is shown you how lacking in evidence Bartemis has. When he can provide corroborating EVIDENCE then we can continue this discussion. You seem to think that I have something to prove. You are wrong. All I have done is poke holes in Bartemis’s pathetic little “theory.” If you wish to support his point of view, please provide EVIDENCE to support him. Otherwise just go away.
Bartemis says ” I, on the other hand, can derive it all from scratch”
…
Too funny…..
…
“Deriving” something is not “evidence”
..
You don’t even know the first thing about science.
..
Show us the DATA
..
Unless you are living on the planet Zircon, science is based on DATA, not derivations
Look Bartemis, your sophmoric “theory” that dCO2/dt = f(T) is pointles for the following reasons…
..
1) It offers no predictive value.
2) It does not explain why there has been a 0.7 degree rise in global temps in the past 150 years
3) It is based entirely on correlation, and not on established physics.
4) It does not fit with established observations of the ice core record.
5) It does not fit with established observations of the concentrations of carbon isotope observations.
6) Your “conjecture” has not been published in a reputable scientific journal. Are you having trouble getting it past peer review?
“1) It offers no predictive value.”
It does. If temperatures remain in stasis, or start to decline, then the rate of change of CO2 will, too. We have already seen this. With the onset of the “pause”, the rate of change of CO2 became essentially constant, while emissions kept accelerating. That in itself falsifies the human attribution hypothesis.
It is worth noting that, it also falsifies the Karlization of global temperatures, as the rate of change of CO2 gives another measure of the temperature level, and it did, indeed, plateau over the past two decades.
“2) It does not explain why there has been a 0.7 degree rise in global temps in the past 150 years”
So what? Neither does the hypothesis of CO2 induced warming, as we see that CO2 does not drive temperature, but rather the reverse. If there were both a significant positive response to temperature from CO2, and this positive integral response to CO2 from temperature, there would be an unstabilizable positive feedback loop, which would have sent us into a runaway greenhouse eons ago. AGW is kaput.
“3) It is based entirely on correlation, and not on established physics.”
It is based on commonly occurring physical relationships. Better than being based on cartoon physics – back-of-the-envelope simple relationships which do not hold in a complex world of many stabilizing feedbacks – such as the failed AGW hypothesis.
“4) It does not fit with established observations of the ice core record.”
Observations which are unverifiable by any independent means. The history of such unconfirmed relationships holding up under scrutiny is rather bleak, once people get past the dogma and start scrutinizing.
“5) It does not fit with established observations of the concentrations of carbon isotope observations.”
Only simple minded interpretations of those observations. But, diffusion processes are complex, and often anti-intuitive.
“6) Your “conjecture” has not been published in a reputable scientific journal. Are you having trouble getting it past peer review?”
This isn’t my job. But, you don’t have to be a weatherman to know which way the wind is blowing. Once the temperatures turn around, the dogma loses currency, and the corrupted pal-review embargo is lifted, I expect you will see a storm of literature addressing the data, instead of sweeping it under the rug.
Clyde Spencer says:
I’m aware of the claim that ice core data supports the claim that CO2 levels followed the increase in temperature. However, from my reading it appears to be a contentious claim.
It’s not contentious, it is completely one-sided. There are numerous data-based charts showing cause and effect between CO2 and temperature. Every one of them shows that ∆CO2 is caused by ∆temperature. There may be others that show ∆CO2 is the cause of changes in temperature, but I haven’t seen them. Whenever I ask anyone to post such a chart, the only ones ever posted are simple overlay charts. But an overlay does not show causation. So Clyde, if you have any charts showing that changes in CO2 cause subsequent changes in global T, please post them. I am sincerely interested.
Janice Moore says:
Bottom line: CK, the burden of proof is on you and all the other AGWers to provide even ONE quantitative measurement that makes the conjecture about human CO2 emissions rise even to the level of a valid hypothesis. So far, not one piece of data proves causation. The climate models’ projections are anti-proof, they failed so badly. And even CORRELATION is against you: CO2 UP. WARMING STOPPED.
The basis of the entire global warming debate from the get-go has been about the CO2=cAGW conjecture. Janice is right; skeptics have nothing to prove. The onus is entirely on those sounding the ‘dangerous manmade global warming’ alarm. But all available evidence indicates it is a false alarm.
Despite a steady increase in (harmless, beneficial) CO2, global temperatures are not accelerating upward, as endlessly predicted.
There is one requirement necessary for any conjecture, hypothesis, theory or law: they must be able to make accurate, repeated predictions. Otherwise, they are wrong.
But not one multi-million dollar GCM, nor one climatologist or organization was able to predict the most significant global temperature event of the past century: the fact that global warming has been stopped for close to twenty years.
Thus, the CO2=cAGW conjecture is wrong. QED
At this point, the honest thing to for those promoting that conjecture to do is to admit that it has failed to predict reality, and to try to formulate a hypothesis that works better at explaining the real world. But instead, the alarmist crowd digs in its collective heels and argues incessantly, trying to support their failed conjecture.
“If an honest man is wrong, after demonstrating that he is wrong, he either stops being wrong or he stops being honest.”
The alarmist crowd has stopped being honest.
db,
if you were to ask Mosher or Stokes about the cause and effect, I don’t think that they would agree with you. Therefore, I think I’m justified in saying the issue is contentious. Personally, I think the issue is still unresolved to everyone’s satisfaction. I can’t give you any chart that resolves the issue. I tried to avoid getting into cause and effect, instead relying on correlation to suggest a plausible extrapolation. It is probably a case of both sides being right because CO2 is part of a feedback loop. I’m trying to keep my mind open to the subject and I’m always looking for the ‘silver bullet.’ One commenter pointed out that CO2 appears to have leveled off during the recent ‘hiatus.’ However, that’s an almost immediate response, not an 800-year delay. How would you interpret an immediate interaction between CO2 and temperature?
Clyde Spencer says:
if you were to ask Mosher or Stokes about the cause and effect, I don’t think that they would agree with you.
That’s OK, they don’t agree with Planet Earth, either. And their side makes it contentious, not skeptics of the ‘dangerous manmade global warming’ conjecture.
And:
How would you interpret an immediate interaction between CO2 and temperature?
I’m not interpreting, I just look at the alarmist conjecture and see one great big FAIL.
The basic debate is over the repeatedly falsified claim that a rise in CO2 will have bad effects. That has not happened. Everything we observe demonstrates that the rise in CO2 has been completely harmless, and greatly beneficial to the biosphere. The earth is measurably greening as a result of more CO2 (which anyway has risen by only one part in 10,000 — over a century and a half). There has been no global damage or harm from the added CO2. Thus, the rise in CO2 is “harmless”. On any cost/benefit analysis, the rise in CO2 has been a net benefit. What is the downside? Can you quantify anything?
The promoters of the CO2=AGW conjecture have failed to make their case. None of their scary predictions have happened, not a single one. In any of the hard sciences, such an abysmal failure would mean the end of it, along with a cut off of grants. The only reason that hasn’t happened here is because the ‘carbon’ scare has become entirely political, with only a thin veneer of science to try and make it appear plausible.
But it isn’t. That conjecture is a complete scientific failure. So now it’s all politics, based on pseudo-science. Wouldn’t you agree?
Dear Janice,
I know, you are a big fan of Dr. Salby and Bart, but you are wrong about who supports who: on the same side of my stance that humans are responsible for the increase in CO2 are a lot of distinguished AGW critics:
Dr. Singer, the Nestor of all skeptics,
Dr. Lindzen,
RGB, who unfortunately did give up to comment here because too much noise of unconvincible people…
Willis Eschenbach
and many others…
I could even convince Dr. Spencer, after an extensive discussion…
Do we need to go through the list, from helicobacter pylori to continental drift and many others, of the times that the majority of scientists were opposed by a single gifted person, and that person was proved right?
Bart,
I was reacting on the claim from Janice that my position was only supported by AGW people, which isn’t true. But I agree, there is no reason that the number of supporters says anything about the truth of a theory.
But violating even one observation does refute even the nicest theory. As yours violates about all known observations, I shouldn’t claim that my theory is of any value and defend it against all odds. No matter how brilliant the inventor of that theory…
Keating says:
Deriving something is not evidence.
But when Clyde Spencer says:
I tried to avoid getting into cause and effect, instead relying on correlation to suggest a plausible extrapolation. I can’t give you any chart that resolves the issue.
Then that argument fails. All it is based on are admissions, like this:
I can’t give you any chart that resolves the issue.
I’ve posted charts showing cause and effect. Clyde admits he can’t do that:
I tried to avoid getting into cause and effect, instead relying on correlation to suggest a plausible extrapolation.
Compared with causation, correlation fails.
Finally, this whole issue is contentious only because the alarmist crowd refuses to follow the Scientific Method. Skeptics are the only honest kind of scientists, but there aren’t any skeptics among the alarmist contingent.
Without skepticism, it’s just politics.
Dbstealey says: “I’ve posted charts showing cause and effect.”
..
No, all you have posted are overlay charts. All they show is correlations.
“Another anthropogenic effect is Urban Heat Island contamination of the temperature records as the cities have encroached on what were formerly rural areas. The BEST project claims to have disproved that hypothesis, but it is my opinion that they didn’t search far enough outside the city limits, nor in the right direction. Quattrochi et al. (Project Atlanta, 1999) have demonstrated that the heat and pollution from central Atlanta (GA) influences the weather for miles downwind from the city. ”
1. our search had nothing to do with CITY LIMITS.
2. We categorized stations using several criteria. non urban was defined as
A) NO “built” areas as defined by MODIS within 10km of the station
B) NO built areas as defined by modis within 20KM of the station.
C) No built pixels within 10km per modis, no impervious surface within 10 km per Imhoff,
No Nightlights within 10km, Population less than 10 people per sq km within 10km.
Basically I looked at every indicator I could, not just modis
3. The Trend in urban versus non urban was INDISTINGUISHABLE. That is, you could not say
that they were different. That doesnt mean UHI is NON EXISTENT. quite the opposite
It means that the UHI we can see at individual sites is not pronounced enough to rise
above the noise in a global record.
4. The atlanta study looked at LAND SURFACE temperatures, for 19 days of clear sky weather.
Do you know why? because UHI is the worst in clear sky conditions. Note that Land surface temperature is different from Surface AIR temperature. Note also that the typical LAND SURFACE
air temperature measurement has an accuracy of +- 1C. Why? because the sensor doesnt
measure the surface. It measures the radiation that comes from the surface and you have to
account for transmission through the atmosphere… With a MODEL.. The better study here
is Pengs study of OVER 400 large cities.. Not just atlanta. read more.
here is a quick over view
https://wiki.lsce.ipsl.fr/pku/lib/exe/fetch.php?media=blog:abstract_presentation1_s_peng.pdf
Full paper
http://pubs.acs.org/doi/abs/10.1021/es2030438
Like most modern studies there is NO CONSISTENT RELATIONSHIP between population density and
UHI. The variables that do matter in a consistent fashion are outlined in Pengs paper. You should read it.
Why is UHI so hard to pin down in the GLOBAL record?
Landform will mitigate or REVERSE UHI.
http://link.springer.com/article/10.1007%2Fs00704-015-1577-9
OR the MICROSITE of an urban station can lead to NEGATIVE UHI
http://nargeo.geo.uni.lodz.pl/~icuc5/text/O_3_1.pdf
“On the basis of six-year-automatic measurements (1997-2002) of automatic measurements in Lodz an empirical studies on the intensity of the urban heat island (UHI) were provided. Continues observations made every 10 minutes allowed to analyze the temperature contrasts between the urban and rural sites in detail. Results show that a typical course of UHI normalized according to sunshine and sunrise time is very similar in all seasons.Comparison of daily courses of the urban-rural temperature contrasts for two sites located in the town center shows that a negative UHI in the morning might be attributed to the microclimatical properties of the site rather than to the physical properties of the whole city. ”
This much can be said. Skeptics had a hypothesis: If you only look at rural stations, the warming
will disappear. Well, we did that. We still found warming. Calling Dr. Feynman…….
Calling karl Popper….
Steven,
Thank you for the links. They were interesting reading. One of the more interesting graphs was the one showing the urban and rural populations intersecting around 2010. I don’t think that you have disproved my claim. The paper showed consistently higher urban temperatures than rural temperatures. So, as a city builds out, it drags its high temperatures into what had formerly been rural land. Should one not see a large increase in the temperatures of the converted land-use area? How does that weight the average temperatures? Also, did you specifically examine and compare rural areas upwind and downwind from large cities? I’m reminded of a complaint a former colleague made about the Three Mile Island event. The NRC said that the average dose received by people living in the area defined by a circle of a given radius did not receive a significant average dose of radiation. What didn’t get said was that in a plume downwind from the reactor, the people DID receive a significant dose. Something that the paper didn’t explain was that there appeared to be irregular cooling at night in Beijing. When you did your analysis of urban vs. rural, did you compare night-time temperatures downwind from cities?
Keating says:
Skeptics had a hypothesis
Wrong from the get-go, sockpuppet. Skeptics simply say:
‘Prove it. Or, provide credible evidence to support yout conjecture’.
You failed.
Next…
If you are going to try to produce a study that is at least somewhat scientifically usable then the equations you come up with should have some meaning. The two regression equations in Fig. 3&4 are given with meaningless coefficients of 0.000. Either choose a proper scale or do not bother wasting a reader’s time with meaningless results. This type of stuff is frustrating for a scientist to see.
Max,
Tell me how you intend to use the negligible coefficients and I will be happy to provide them.
Clyde Spencer,
As Steve Mosher already said: too many errors in this story, so where to start in my comment?
1. To start with the obvious: a one-on-one correlation in a multivariate system seldom gives a good answer. In this case there is one source of CO2: human emissions with a huge trend and little variability and a second source: temperature with a small trend and huge variability. The accumulation of CO2 emissions by humans, temperature and the increase in the atmosphere can be plotted together:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_emiss_increase.jpg
Which already shows that the correlation between temperature and CO2 is not that good, but both CO2 emissions and CO2 increase in the atmosphere show a similar slightly quadratic increase over time.
CO2 emissions are based on fossil fuel sales and burning efficiency and CO2 levels in the atmosphere are based on the 10-year resolution of the Law Dome ice core until 1959 and Mauna Loa since 1960.
The slopes are a near fit:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1900_cur.jpg
Since 1960:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/acc_co2_1960_cur.jpg
There is an obvious possibility of a causal origin of the increase in the atmosphere, which may be correlated to both human population and its wealth: increasing population and increasing energy use, thus CO2 emissions.
Compared to the influence of temperature (either way):
http://www.ferdinand-engelbeen.be/klimaat/klim_img/temp_co2.jpg
A change of half the scale has little influence on CO2 increase, but the full trend would give a lot of CO2? Not really convincing…
2. Historical CO2 levels and temperatures in ice cores show a change of ~16 ppmv/°C over the past 800,000 years. That coincides with Henry’s law which gives ~16 ppmv/°C for the equilibrium between ocean surface waters and the atmosphere. Thus the ~1°C warming since the LIA is good for maximum 16 ppmv increase, the rest is from human emissions…
3. OCO-2 shows the difference between equator and poles: it doesn’t show the net effect. There is a continuous flux of CO2 of about 40 GtC/year between the upwelling waters near the equator and the sinking waters near the poles. Based on pCO2 measurements of waters all over the world, the oceans are more sink than source…
That the source of the extra CO2 is surely in the NH can be seen if you compare the yearly averages over time:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/co2_trends_1995_2004.jpg
4. Human induced water vapor is of no influence at all: I have calculated it some long time ago and it is about 0.01% of the gigantic water cycle… Compare that to CO2 emissions, which are already around 6% of the natural cycle…
I’ll have to agree with both Ferdinand and Mosher. Perhaps Clyde can rework it to fix the errors.
Mike,
What evidence? The 16 ppmv/°C is what Henry’s law says for the solubility of CO2 in seawater. In the literature that varies between 4-17 ppmv/°C. As vegetation is proven (small) net sink for CO2, the only non-human source of interest are the oceans, all other possible sources (volcanoes, rock weathering,…) being too small or too slow…
Confirmed by the CO2-temperature (proxy) ratio in ice cores.
Confirmed by over 3 million seawater samples over the past decades…
Sorry, wrong place for this reaction, should be downwards…
Clyde,
The deep oceans are indeed lower in δ13C than the ocean surface by the exchange with dead plant and animal material, but the deep oceans still are around zero per mil, while the surface is at +1 to +5 per mil, depending of the abundance of bio-life in the surface layer which is very rich in the upwelling zones…
Nevertheless, there is an isotopic shift in δ13C at the water-air border and reverse, due to the slower speed of escape of 13CO2 for its larger mass. That makes that at the long time equilibrium between deep and surface oceans and vegetation, the atmosphere was at -6.4 +/- 0.2 per mil over the full Holocene before 1850.
Ferdinand,
So let me see if I understand the problem, at least at a qualitative level. It is believed that the rise in atmospheric CO2 is principally anthropogenic because fossil fuels are low in C13, and the burning of fossil fuels primarily add C12 to the atmosphere. However, living organisms selectively use C12 over C13. When the plankton die, and are consumed in their downward travel through the water column, their decomposition results in the creation of C13-low CO2. When that CO2 is released by upwelling, it is already low in C13 and is further depleted in C13 by selective evaporation. This seems like circular reasoning to use the low C13 as evidence that it is coming from fossil fuels. If the oceans are warming, which they seem to be, then there should be an increased rate of outgassing over what happened in the past and there should be an enrichment of C12. Did I get this right?
Clyde:
When the plankton die, and are consumed in their downward travel through the water column, their decomposition results in the creation of C13-low CO2. When that CO2 is released by upwelling, it is already low in C13 and is further depleted in C13 by selective evaporation.
The organics of dead plankton sinks into the deep oceans and are decomposed by bacteria. While this resulting CO2/derivatives is low in δ13C, maybe -20 per mil or so, that is getting into the bulk of mainly inorganic carbon. The gigantic amount of CO2/bi/carbonate in the deep oceans is around zero per mil and is hardly influenced by the yearly drop out of organics from the surface, the more that also carbonates are dropping out from the shells (and added from dissolved sediments on land) and everything seems to be equilibrated over many millennia.
When the oceans and the atmosphere exchange CO2, there is indeed a shift in isotopic ratio: -10 per mil between oceans and atmosphere, +2 per mil between atmosphere and oceans, both seen from the atmosphere. Thus average -8 per mil change for any continuous flux of CO2 between oceans and atmosphere and back.
If all fluxes were with the deep oceans, that would give an equilibrium of around -8 per mil in the atmosphere. But there is abundant bio-life in the surface, especially at the upwelling zones. That makes that the release of CO2 is from waters slightly above zero per mil δ13C and the average in the atmosphere over the whole Holocene is -6.4 +/- 0.2 per mil… until humans became involved…
The δ13C level of the atmosphere since ~1850 dropped to below -8 per mil. If there was a substantial increase in the oceanic-atmospheric carbon cycle, that would increase the per mil of the atmosphere. If it was a one-way input, that would add CO2 of somewhere around -8.5 per mil, a little below current levels.
But the average fluxes, based on pCO2 differences, give more sink than source (~3.5 GtC/year), thus should increase the δ13C level in the atmosphere…
This is a better fit:
http://i1136.photobucket.com/albums/n488/Bartemis/tempco2_zps55644e9e.jpg
Bart,
Integration of temperature has no bearing in any physical process: if the temperature shows a step increase, CO2 doesn’t accumulate until eternity: with 16 ppmv/°C, a new equilibrium is reached, no matter if that is static in a bottle or dynamic all over the ocean sources and sinks.
The increase of 0.7°C since 1959 is good for 11 ppmv increase of the 90 ppmv increase that is all…
Nonsense. I have shown mathematically how such a relationship can arise.
This is a continuous flow problem. CO2 is always entering the surface system. If it is impeded from exiting, e.g., by an increase in temperature, it will accumulate in the surface system, giving rise to a sustained increase.
Here is a more detailed mathematical model.
Ferdi, then clearly your assertion that “… is good for 11 ppmv increase ” does not fit the evidence. You have been wheeling this figure out of years.
Bart,
Mathematically one can prove that any mixture of temperature and human emissions can fit the total curve of slope and variability, but again, that needs to be based on physical processes.
The physical process behind the variability is mainly the influence of temperature (and rain patterns) on tropical vegetation The physical process that gives the slope is not vegetation. Neither the oceans, as both are proven net sinks for CO2…
You have zero proof that CO2 accumulates in the ocean sinks, that is just your imagination to save your theory…
Your dCO2/dt = k*(T – T0) lacks any physical response of the ocean in and out fluxes to the increasing CO2 pressure in the atmosphere. The fluxes either way are directly proportional to the pCO2 difference between the ocean surface and the atmosphere. If the temperature of the ocean surface increases, its pCO2 increases with 16 μatm/°C. As result, the atmosphere increases its CO2 content and with 16 μatm/°C (~16 ppmv) extra in the atmosphere, the original in and out fluxes are fully restored and no further increase of CO2 will happen.
That is where your mathematical model goes wrong: you didn’t take into account the reaction of the fluxes on the increased pressure in the atmosphere…
“…that needs to be based on physical processes.”
It is. It is a model of the atmospheric/oceanic equilibration process, and its temperature sensitivity.
“The physical process behind the variability is mainly the influence of temperature (and rain patterns) on tropical vegetation…”
Assertion.
“The physical process that gives the slope is not vegetation.”
Assertion.
“Neither the oceans, as both are proven net sinks for CO2…”
Assertion based upon circular logic.
“You have zero proof that CO2 accumulates in the ocean sinks…”
I have proof that temperature drives the rate of change of CO2. The ocean exchange mechanism is a viable process for causing that.
“…you didn’t take into account the reaction of the fluxes on the increased pressure in the atmosphere…”
I very specifically did. It’s all right there in the math.
Bart,
Where in your formula dCO2/dt = k*(T – T0) is the influence of the increased CO2 pressure in the atmosphere?
In that formula the only influence on the CO2 rate of change is the temperature difference with an arbitrary start temperature. Even if the CO2 pressure in the atmosphere increases to far above the pCO2 everywhere in the ocean surface, there still gets more CO2 into the atmosphere without any temperature increase, only a small offset is sufficient…
The formula is derived from a model which includes the equilibration process between the oceans and atmosphere here.
You are arguing incoherently. CO2 accumulates in the surface system due to temperature induced throttling of its transport via the THC. That represents a net input, every bit as much as human inputs represent a net input, to the surface system. The only difference is that human inputs are comparatively small. If you argue that, that accumulation cannot affect atmospheric concentration, then you are also arguing that human inputs cannot, either.
Bart,
That is a new one, which uses so many variables and assumptions that you can “prove” anything you want…
Simply said: if the net result of all your assumptions is that the increase in the atmosphere is only dependent of a (small) offset in temperature without any negative feedback from the increased pressure in the atmosphere, then at least one of the assumptions is completely wrong…
CO2 accumulates in the surface system due to temperature induced throttling of its transport via the THC.
Say what? Temperature has not the slightest influence on CO2 levels once it is in the waters and hardly any on the water flow itself: the main sink place follows the edge of the ice where temperature is near zero – or below and the density gets high enough to sink the waters. There is no “throttling” of CO2, as these cold waters are always highly undersaturated in CO2, compared to the atmosphere.
All what can happen is that an increased temperature of the surface increases its pCO2 and for the same pressure in the atmosphere, less is absorbed. The opposite happens at the upwelling places near the equator: any temperature increase will increase the pCO2 of the ocean surface and thus increase the influx. Both increase the CO2 levels in the atmosphere, but the original fluxes are fully restored at 16 ppmv/°C increase in the atmosphere…
BTW, my “assertions” are based on field measurements, your assumptions are not based on any observations…
“…without any negative feedback from the increased pressure in the atmosphere…”
All the feedbacks are in the math. This is how the math works out. Your intuition is not up to this task.
“Temperature has not the slightest influence on CO2 levels once it is in the waters…”
Temperature very much determines how much CO2 remains in the waters to downwell. Where it is upwelling, the content is essentially an exogenous input, not influenced by current conditions. So, you have a more or less constant input rate, but a temperature variable rate of output.
When temperature increases above an equilibrium level, you will get a steady trend upward. When it falls below that level, you will get a steady trend downward. It cannot go on forever – at the very worst, if temperatures remain consistently elevated or depressed, it will only last for as long as it takes for full turnover, which is about 800 years or so. But, over relatively short timelines, it will appear to track the integral of temperature.
“BTW, my “assertions” are based on field measurements, your assumptions are not based on any observations…”
Your assertions are based on a particular reading of the observations. My assumptions are based on the fact that the rate of change of atmospheric CO2 tracks temperature anomaly.
Bart,
All the feedbacks are in the math. This is how the math works out. Your intuition is not up to this task.
If the end result of all your math shows that 1 + 2 = 4, I don’t need to even follow all what you have done to know that one of your assumptions must be wrong as your conclusion is mathematically impossible
If the end result of all your assumptions is that dCO2/dt = k(T – To) then I don’t even need to follow all your many steps to know that one of your assumptions must be wrong, as your formula doesn’t show any influence of the increased CO2 pressure in the atmosphere on the CO2 rate of change. That is physically impossible.
Temperature very much determines how much CO2 remains in the waters to downwell.
Temperature determines the local pCO2 pressure of the ocean surface waters. It determines the pCO2 difference between atmosphere and ocean surface and that determines the CO2 flux from the atmosphere into the downwelling waters, as at the sink places the pCO2 of the cold polar water is much lower (~250 μatm) than in the atmosphere (~400 μatm), a difference of ~150 μatm. The outflux is directly proportional to the pCO2 difference.
If the local surface temperature increases with 1°C then the local surface pCO2 increases with 16 μatm and the CO2 outflux into the downwelling waters is reduced to (150-16)/150 = 89% of the original outflux.
That increases the CO2 content of he atmosphere – all other in/outs remaining equal. The increase of CO2 in the atmosphere again pushes more CO2 into the downwelling and with 16 ppmv extra CO2 pressure in the atmosphere, the original outflux is fully restored…
The influence of the higher CO2 pressure from the increase in the atmosphere on the outflux is exactly what lacks in your formula…
Where it is upwelling, the content is essentially an exogenous input, not influenced by current conditions.
Sorry, can’t be true: whatever the carbon content of the deep ocean upwelling, its CO2 release only depends of the local pCO2 difference between ocean surface (~650 μatm) and atmosphere (~400 μatm), a difference of ~250 μatm.
For a fixed upwelling (both in water amount and carbon concentration), temperature plays its role, again with ~16 μatm/°C. For 1°C temperature increase, the influx increases to (250+16)/250 = 106% of the original influx. Again that increases the CO2 content in the atmosphere. Again the increased CO2 pressure in the atmosphere reduces the CO2 influx from the upwelling zones until at 16 ppmv extra in the atmosphere the original influx is fully restored…
The influence of the higher CO2 pressure from the increase in the atmosphere on the influx is exactly what lacks in your formula…
When temperature increases above an equilibrium level, you will get a steady trend upward
Which is physically impossible. That violates Henry’s law for the solubility of CO2 in seawater at different temperatures. That ignores the influence of the increased CO2 pressure in the atmosphere which is already 110 ppmv above the dynamic equilibrium for the current ocean surface temperature.
Your assertions are based on a particular reading of the observations.
Like that the firm decrease in 13C/12C ratio in the atmosphere is impossible from oceanic CO2 releases which have a much higher 13C/12C ratio?
“The increase of CO2 in the atmosphere again pushes more CO2 into the downwelling and with 16 ppmv extra CO2 pressure in the atmosphere, the original outflux is fully restored…”
You are forgetting that, every second during which that atmospheric CO2 is increasing, CO2 is piling up in the surface waters, further increasing the amount by which the atmospheric concentration has to increase to push it back down. By the time you have increased 16 ppmv, the concentration of the ocean has increased so that you have to increase to 32, then 48, then on and on and on.
It’s a dynamic system, such as you have so much trouble grasping, Ferdinand. Nothing is staying the same. Your static accounting is simply not up to the task. Remember this?
https://youtu.be/8NPzLBSBzPI
“…whatever the carbon content of the deep ocean upwelling, its CO2 release only depends of the local pCO2…”
And, the local pCO2 is continually piling up, as downwelling is restricted.
“That violates Henry’s law for the solubility of CO2 in seawater at different temperatures.”
It doesn’t. This is not a static situation. It is not a pool of static CO2 content subjected to a temperature rise. It is a transport phenomenon, in which there is a continual influx of CO2 from below the surface waters.
“Like that the firm decrease in 13C/12C ratio in the atmosphere is impossible from oceanic CO2 releases which have a much higher 13C/12C ratio?”
It’s not impossible, you just don’t know of another source. This is an argument from ignorance, akin to saying that the rumbling volcano must be a sign of the gods’ discontent, because you cannot think of any other possibility.
Moreover, it assumes that the diffusion processes through which the flows circulate are symmetric and homogenous with respect to 12C and 13C. So, at best, you can only say the 13C/12C ratio is consistent with your hypothesis. But, consistency is not proof.
Bart:
You are forgetting that, every second during which that atmospheric CO2 is increasing, CO2 is piling up in the surface waters, further increasing the amount by which the atmospheric concentration has to increase to push it back down.
Bart, you have a remarkable vision of how the CO2 exchange works…
As long as the upwelling waters don’t pile up (which would be a spectacular view), CO2 doesn’t “pile up” in the surface waters: what comes in as upwelling water near the Peruvian/Chilean coasts flows westward, rounds Cape Town and passes the Gulf of Mexico on its way to sink again in the NE Atlantic. Thus while fresh CO2 continuously comes out of the deep, that is within waters that don’t remain on the same spot.
All what happens is that the cold waters from the deep are warmed up to tropical temperatures and thus the pCO2 increases tremendously (as far as not captured by plankton – fish), which causes the huge (~40 GtC/year) continuous influx of CO2. That is all. No reason at all to increase the CO2 releases further, as long as the upwelling water flow and its CO2 content remains the same. Only temperature modulates that somewhat: ~6% more influx for 1°C temperature increase, because of its effect on the pCO2 of the upwelling waters and thus the pCO2 difference with the atmosphere.
During its journey along the equator, most of the time these waters are a net source of CO2. Once in colder waters the pCO2 of the water drops to below the atmospheric pCO2 and the same waters become a sink for CO2. Ultimately at the sink place, where the coldest temperatures are, the sink rate is maximal as the pCO2 difference with the atmosphere is maximal. As long as the Gulfstream/THC doesn’t stop (à la “The Day after Tomorrow”, but then we have real problems, not CO2), again there is no blocking of the CO2 outflux. Every parcel of water passing by and sinking into the deep takes some CO2 out of the air in ratio to the pCO2 difference with the atmosphere. For the same amount of water and the same concentrations, that is the same quantity of CO2, again only modulated by temperature: 10% less uptake for 1°C temperature increase.
The change of in-out fluxes for 1°C temperature increase thus increases the CO2 level in the atmosphere. With only 16 ppmv more CO2 in the atmosphere the pCO2 difference and thus the fluxes are fully restored as well at the source as at the sink side.
The net effect of the whole cycle (nowadays) is ~3 GtC more sink than source.
See the nice compilation of Feely e.a. at:
http://www.pmel.noaa.gov/pubs/outstand/feel2331/mean.shtml
There is no proof whatever that anywhere in the oceans CO2 piles up, which would be visible as an enormous pCO2 increase. The increase of pCO2 of the ocean waters simply follows the atmospheric increase, not reverse, with an average 7 μatm more CO2 pressure in the atmosphere than in the ocean surface,
It’s a dynamic system, such as you have so much trouble grasping
I have the impression that my knowledge of (slow) real world dynamic systems is a little more realistic than yours…
It’s not impossible, you just don’t know of another source.
All I have said is that it is impossible that the oceans are the cause of the 13C/12C ratio decline in the atmosphere. That is as impossible as adding an acid to a solution and expecting that the pH goes up.
Nothing to do with other possible causes which are so hard to find, while the most obvious cause, human emissions, is just before your nose…
Ferdinand,
Perhaps I can shed some light on this. As I understand it from my reading, the reason that upwelling cold water has a lower pH than surface waters is that as the constant rain of dead organic material moves downwards through the water column, it oxidizes or is is consumed by bacteria, releasing carbon dioxide. This is dissolved in the cold water, which is also under high pressure, This depletes the oxygen and increases the bicarbonate. When that cold water upwells in the tropics, the CO2 comes out of solution. This has implications for the carbon isotope ratio because it should be low-13C, being from organic origin.
Clyde,
repeat from above…
The deep oceans are indeed lower in δ13C than the ocean surface by the exchange with dead plant and animal material, but the deep oceans still are around zero per mil, while the surface is at +1 to +5 per mil, depending of the abundance of bio-life in the surface layer which is very rich in the upwelling zones…
Nevertheless, there is an isotopic shift in δ13C at the water-air border and reverse, due to the slower speed of escape of 13CO2 for its larger mass. That makes that at the long time equilibrium between deep and surface oceans and vegetation, the atmosphere was at -6.4 +/- 0.2 per mil over the full Holocene before 1850.
“No reason at all to increase the CO2 releases further, as long as the upwelling water flow and its CO2 content remains the same.“
That is the entire point. The CO2 content of the surface oceans does not stay the same. Less CO2 downwells with the downwelling water. Hence, the surface oceans have rising content, which results in rising content of the atmosphere.
“There is no proof whatever that anywhere in the oceans CO2 piles up, which would be visible as an enormous pCO2 increase.”
It is. The oceanic pCO2 increase is always commensurate with the atmospheric increase, via Henry’s law. Both atmospheric and oceanic CO2 have been increasing. That is the reason some people have been warning of acidifying oceans.
“I have the impression that my knowledge of (slow) real world dynamic systems is a little more realistic than yours…”
You were suckered in by the pseudo-mass balance argument for the same reason. You do not think things through where dynamic processes are involved.
Bart:
That is the entire point. The CO2 content of the surface oceans does not stay the same. Less CO2 downwells with the downwelling water. Hence, the surface oceans have rising content, which results in rising content of the atmosphere.
Higher temperatures give more CO2 release from the oceans at the upwelling side and less uptake at the sink side, in both cases the remaining carbon (DIC: CO2 + derivatives) in the ocean water stream would go down if the oceans were the cause of the atmospheric increase. The rest of the oceans surface outside the THC circulation and its sources and sinks also would have a dropping DIC content from its warming, while it goes up everywhere where was and is measured…
The warming oceans theory thus is opposite to what is measured, not only in content (DIC) but also the average pCO2 of the ocean surface, which is ~7 μatm below the pCO2 of the atmosphere. That means that the net flux is from the atmosphere into the oceans, not reverse. About 3 GtC/year more CO2 is downwelling than upwelling due to the increased CO2 pressure in the atmosphere…
It is. The oceanic pCO2 increase is always commensurate with the atmospheric increase, via Henry’s law. Both atmospheric and oceanic CO2 have been increasing. That is the reason some people have been warning of acidifying oceans.
See previous answer: CO2 content of the oceans follows the atmospheric increase, not reverse. Besides the upwelling places and most of the hot equatorial waters, which are net sources of CO2, the rest of the oceans are net sinks over a full year, even in still warm ocean waters like Bermuda and Hawaii:
http://www.biogeosciences.net/9/2509/2012/bg-9-2509-2012.pdf with the analysis in chapter 4 and especially 4.2 and
http://www.pnas.org/content/106/30/12235.full.pdf Fig. 1
The sink rate is ~0.5 GtC/year in the ocean surface layer and ~3 GtC/year in the deep oceans (total sink in the oceans: 2.2 +/- 0.4 GtC/year in the reference year 1995 according to Feely e.a.).
You were suckered in by the pseudo-mass balance argument for the same reason. You do not think things through where dynamic processes are involved.
The mass balance still must hold in all cases for every second of the day, but indeed there is one escape route for not involving human emissions, which needs a lot of imagination, but still mathematically possible. The problem is that it needs a fourfold increase in the natural cycle to fit the fourfold increase in observed net sink rate, for which is not the slightest indication – to the contrary…
The same here: with a lot of imagination, CO2 may pile up somewhere in the ocean surface, but besides that the surface waters follow – not lead – the increase in the atmosphere, there is no observation anywhere of a huge increase in DIC or pCO2 beyond the increase in the atmosphere…
Before thinking that others don’t think deep enough, look a little deeper into the chemistry and physics of the oceanic CO2 exchanges with the atmosphere, no matter if these are static or dynamic…
“…in both cases the remaining carbon (DIC: CO2 + derivatives) in the ocean water stream would go down if the oceans were the cause of the atmospheric increase.”
No, that is not how it works. In fact, it is the opposite.
When temperatures increase, there is less downwelling CO2. Since there are new CO2 laden waters upwelling all the time, that forces an accumulation in the surface waters. When the surface waters accumulate enough to force it back down in commensurate quantity, the accumulation ceases. But, that takes a very long time.
Outgassing to the atmosphere actually impedes this process slightly, because it provides an alternative path for the CO2 to exit from the oceans, lessening the downward pressure.
The atmosphere is simply along for the ride. The oceans are the elephant, and the atmosphere is the flea on its back.
“CO2 content of the oceans follows the atmospheric increase, not reverse.”
Nonsense. The timeline of equilibration between ocean and atmosphere is fast, and the observations are closely in phase. You cannot impute causality from the observations. There is not enough phase differentiation.
“The mass balance still must hold in all cases for every second of the day…”
A true mass balance, yes. The pseudo-mass balance argument, no. The pseudo-mass balance argument does not differentiate between sink activity due to natural forcing, and sink activity due to anthropogenic forcing. It assumes all sink activity is natural when, in fact, a substantial portion of it is anthropogenically induced.
“The problem is that it needs a fourfold increase in the natural cycle to fit the fourfold increase in observed net sink rate…”
No it doesn’t. It simply requires that the downwelling be throttled to the level that the observed rise is what it is.
“…no matter if these are static or dynamic…”
There, you are totally and completely wrong. There are huge differences between static and dynamic processes. Many folks out there who followed a scientific or engineering curriculum will know well that they generally studied static physics first, before moving on to advanced dynamics classes. You start with trusses, and end with jet engines. The ways you address these problems are very, very different.
Bart:
No, that is not how it works. In fact, it is the opposite.
When temperatures increase, there is less downwelling CO2. Since there are new CO2 laden waters upwelling all the time, that forces an accumulation in the surface waters.
Bart, this is really physically impossible. There is a 20,000 km distance between upwelling and downwelling of mostly the same waters. There is no physical back connection between downwelling waters and the rest of the ocean surface for CO2 via water at all, it is one way from upwelling to downwelling of average the same quantity of water, including its CO2 content.
Any parcel of water flowing in the direction of the sink place doesn’t know what the uptake of CO2 at the sink place is, it only releases or absorbs CO2 in ratio to the pCO2 difference between itself and the atmosphere above it, hardly with the parcel before or behind it.
Thus how and where in your opinion can CO2 in the water phase increase, if not via the atmosphere?
Any CO2 contained in the upwelling waters will be moved either into the atmosphere, based on the local pCO2 difference, or into the sinks at the other end of the world (or captured by plankton). Again there is no way, none at all, that CO2 can pile up in the waters from less downwelling CO2, if not coming out of the atmosphere… The more that during the trip the waters go from hot to cold, thus what was first released is (at least partly) replenished with CO2 out of the same atmosphere.
If there is any CO2 increase in the ocean surface anywhere from less CO2 downwelling, the only possible way is via the atmosphere. And the only forcing in both directions is the pCO2 difference between atmosphere and oceans.
Nonsense. The timeline of equilibration between ocean and atmosphere is fast, and the observations are closely in phase. You cannot impute causality from the observations. There is not enough phase differentiation.
Again, as with temperature, you are looking at the variability of the noise, not the trends. The increase has a “wave” length of over 600 years, if any at al, but you are expecting to find a phase differentiation…
Simply look at the level differences: the pCO2 levels in the ocean surface lag the pCO2 levels in the atmosphere with years from 20N to sea ice in the Arctic and from 20S to sea ice in Antarctica.
Average with a difference of ~7 μatm over the full ocean surface, including the tropics. For an increase of ~2 μatm (~ppmv) per year that is an average lag of ~3.5 years. More than enough to be sure that the net flux is from atmosphere into the ocean surface, not reverse.
No it doesn’t. It simply requires that the downwelling be throttled to the level that the observed rise is what it is.
The “real” mass balance shows that the net sink rate quadrupled over the past 55 years, nothing to do with throttling the downwelling, it is just the opposite.
Any combination of natural and human causes must show a 4-times increased net sink rate.
For human emissions that is no problem: these increased a fourfold in the past 55 years.
The increase rate in the atmosphere also increased a fourfold in the same period.
The difference between both shows a fourfold increase in net sink rate with zero increase in the natural cycle.
If you throttle the sinks to give a fourfold increase in the atmosphere and at the same time human emissions increase a fourfold, where do human emissions get? If natural emissions are the cause, the sinks should have increased a fourfold for natural + human emissions together, as sinks don’t differentiate between the CO2 origin…
There are huge differences between static and dynamic processes
Agreed, but there are also huge differences between the dynamics of radio frequencies or a gas turbine and the slow dynamics of ocean – atmosphere exchanges, which are far closer to pseudo-static (“steady state”) processes than to highly dynamic processes where your experience obviously is…
“There is no physical back connection between downwelling waters and the rest of the ocean surface for CO2 via water at all, it is one way from upwelling to downwelling of average the same quantity of water, including its CO2 content.”
Nonsense. CO2 diffuses and advects from the areas of upwelling to the areas of downwelling. Temperatures play a major role in that diffusion and advection.
This is a transport phenomenon. Delays set up oscillatory modes of transmission and reflection. On average, the quantity of waters upwelling and downwelling must balance. But, they will not generally balance instantaneously, and the solution content of those waters is not generally going to be the same at all.
You are rationalizing, waving your arms all around, and insisting that the system must unfold according to your static rules. It is absurd.
“The “real” mass balance shows that the net sink rate quadrupled over the past 55 years, nothing to do with throttling the downwelling, it is just the opposite.”
It just shows that the balance between sink and source has been shifted. You are getting wrapped up in rationalizations to preserve your conception. It is not unlike the epicycles used to justify variations in planetary orbits in the vain attempt to preserve the geocentric universe. You are fooling yourself.
“…and the slow dynamics of ocean – atmosphere exchanges, which are far closer to pseudo-static (“steady state”) processes than to highly dynamic processes where your experience obviously is…”
“Slow” dynamics are just fast dynamics with a different time scaling. Thread them through the projector and increase the frame rate, and they become fast. It is a matter of perception, not of reality.
‘“Slow” dynamics are just fast dynamics with a different time scaling.’
Let me elaborate on that a bit. Time is always relative. When you say something is “fast”, you are implicitly comparing it to something that is slower. Without the comparison, the concept of “fast” is meaningless.
This is why newbies always have such a hard time grasping relativistic concepts. We are hard-wired by experience to percieve time relative to processes with which we are familiar, and to imagine that it must always unfold at the same rate. But, that is simply not the case. Time is the rate at which processes evolve relative to other processes. And, time in a frame of reference accelerated away from another frame of reference evolves more slowly relative to the original frame of reference.
So, when you say a process is “slow”, you are implicitly comparing it to your familiar experience. But, for a dynamic system, your familiar experience is not an appropriate standard. A system is “fast” if it is evolving rapidly with respect to the natural time constants dictated by the boundary conditions of the system. The natural time constants of planetary scale systems are exceedingly slow, hence, almost any process unfolding relative to them is “fast”.
You can only treat a system as “quasi-static” if the natural time constants are fast relative to the dynamics you are interested in. Otherwise, you are making an unwarranted assumption which is going to produce the wrong answer.
Bart:
Nonsense. CO2 diffuses and advects from the areas of upwelling to the areas of downwelling. Temperatures play a major role in that diffusion and advection.
Migration of salts, temperature, CO2 or anything else that is in the waters is extremely slow. Slower than the speed of the THC/Gulf Stream circulation around the earth. One can have a sudden change of several degrees in temperature, changes in salt content, pH, DIC within a km if you navigate in and out the Gulf Stream. The main water flow is only becoming diffuse in the far NE Atlantic near the sink area.
No matter that point, the main point is that there is no back-propagation of CO2 in the waters from the sink places to anywhere in the ocean surface, not even 100 meters back from the first sinks. The only possibility of “piling up” of CO2 in the surface waters is via the atmosphere. The main changes in CO2 content underway are the exchanges with the atmosphere due to temperature and downward via bio-life. Hardly by mixing with the waters below, before, behind and somewhat more besides.
On average, the quantity of waters upwelling and downwelling must balance. But, they will not generally balance instantaneously, and the solution content of those waters is not generally going to be the same at all.
Doesn’t matter: the quantity of water must be balanced between upwelling and downwelling if averaged over sufficient periods. Any temporary slowdown and speed up will have a temporary effect on temperature, evaporation and CO2 levels via releases and uptake…
In average for a certain surface temperature and a constant CO2 content in a constant upwelling, the pCO2 of the atmosphere will equilibrate at a certain level, where at steady state CO2 influxes and outfluxes are equal, if averaged over sufficient periods. The same for output concentrations.
If the temperature of all ocean surfaces increases, the CO2 influx will increase at the upwelling zones and all warmer zones and the CO2 outflux will decrease at all cold and sink zones. That will lead to an increase of CO2 in the atmosphere, which equilibrates again at ~16 ppmv/°C. Then the CO2 increase stops, as influx and outflux are back to their original strength:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/upwelling_temp.jpg
It just shows that the balance between sink and source has been shifted. You are getting wrapped up in rationalizations to preserve your conception.
Come on Bart, you can do better: show me the math where a combination of natural inputs + a fourfold increase in human inputs minus total (natural + human) outputs causes a fourfold increase in the atmosphere and a fourfold increase in net sink rate…
You are evading that question already for years now…
Thanks for the lesson in dynamic processes, which was a long time ago. Still that doesn’t change the heart of the matter: No matter if the process is static or dynamic:
The equilibrium between ocean surfaces and atmosphere changes with about ~16 ppmv/°C.
And the main physical impossibility:
There is no “piling up” of CO2 in the surface waters by back propagation via the water phase.
“No matter that point, the main point is that there is no back-propagation of CO2 in the waters from the sink places to anywhere in the ocean surface, not even 100 meters back from the first sinks.”
Ridiculous. There is less CO2 in the waters where they downwell, hence less gets transported down, hence it stays in the surface oceans and accumulates.
It has to downwell, Ferdinand, in the same measure as it is upwelling, to achieve balance. If it doesn’t, it’s got to accumulate in the surface system. There is no other place it can go in substantial quantity, and it cannot just wink out of existence.
“…show me the math where a combination of natural inputs + a fourfold increase in human inputs minus total (natural + human) outputs causes a fourfold increase in the atmosphere and a fourfold increase in net sink rate…”
I showed you a mathematical model here.
“You are evading that question already for years now…”
Because it is a flawed question. It is an inaccurate description of what must happen. All that is required is that the oceans absorb most of the CO2 in the system, and temperatures throttled the downflow such that it accumulated in the surface oceans, and thence in the atmosphere.
Bart:
Ridiculous. There is less CO2 in the waters where they downwell, hence less gets transported down, hence it stays in the surface oceans and accumulates.
Bart, you do know a lot about high frequency processes, but you have no idea where you are talking about in this case.
What you say implies that the water flow goes down undisturbed, but (part of) its CO2 content remains in the waters coming in behind it. Would be a little difficult as migration of CO2 in water is extremely slow…
Reality is quite different:
In the upwelling zones, with a certain influx of water with CO2/bi/carbonate at a certain concentration, a fixed amount of CO2 is released into the atmosphere for a fixed temperature over the upwelling and warmer zones in ratio to and as long as the pCO2 of the water is above the pCO2 of the atmosphere.
On its way to the sink places, the waters cool down and start to absorb CO2 out of the atmosphere again in ratio to the pCO2 difference with the atmosphere.
If the oceans surface warms up everywhere, more CO2 is released in the upwelling and warm zones, as the pCO2 of the oceans increased and thus the pCO2 difference with the atmosphere increased. More release of CO2 means that the waters are already more depleted of CO2 when they reach the colder zones.
In the colder zones less CO2 is absorbed by the waters, as the pCO2 of the oceans is increased and thus the pCO2 difference with the atmosphere decreased. Thus a lot less CO2 is going into the deep.
It doesn’t matter how much CO2 was in the waters just before downwelling, as that simply follows the water stream and hardly exchanges its CO2 content with the surrounding waters. In all cases there is no increase of CO2 in water in the whole trajectory, compared to before the warming, only a decrease everywhere caused by the higher temperatures.
There is no piling up of CO2 anywhere in the water phase
Where then is the piling up? Because of more CO2 release in the warmer zones and less absorption in the colder zones, the increase is in the atmosphere, not in the ocean waters, which get depleted. But as release and absorption is directly proportional to the pCO2 difference between atmosphere and ocean surface, the original in/out fluxes are fully restored with an increase of 16 ppmv/°C for any ocean surface warming.
BTW, the uptake capacity of the ocean surface is limited by its buffer factor which makes that a 100% CO2 change in the atmosphere results in only a 10% CO2 (DIC) change in the ocean surface. The part in direct contact (the “mixed layer”) with the atmosphere contains about 1,000 GtC. The atmosphere about 800 GtC. If CO2 piles up, it is in the atmosphere, hardly in the ocean surface…
Because it is a flawed question. It is an inaccurate description of what must happen. All that is required is that the oceans absorb most of the CO2 in the system
Sorry Bart, no matter the processes behind the removal of any extra CO2 in the atmosphere, the observations show a fourfold increase in net sink rate over the past 55 years. So do human emissions and the observed increase in the atmosphere.
There are only two possibilities: either the natural carbon cycle didn’t change one bit and all increase is human or the natural cycle increased a fourfold in the same period in exact ratio to human emissions, because the sinks don’t discriminate between human and natural CO2. That means a fourfold increase in both release and absorption, where the latter is contradictory to any “throttling” of the downwelling…
Ferdinand,
In your exchanges with Bart, it strikes me that you have simplified things a little too much. Strictly speaking, Henry’s Law only applies to relatively inert gases like oxygen and nitrogen. Carbon dioxide, per se, has a solubility that varies with temperature and pressure, but it also chemically interacts with the water to produce carbonic acid (which promptly dissociates), and leaves carbonates and bicarbonates, both of which have different solubilities dependent on temperature and pH. If you go back to the post I made some months ago ( http://wattsupwiththat.com/2015/09/15/are-the-oceans-becoming-more-acidic/ ) you can see that there is little relation between the atmospheric CO2 and pH. Things are much more complex in the real world.
Ferdinand,
In reply to Bart you said, ” In all cases there is no increase of CO2 in water in the whole trajectory,…” That is not a true statement. As we have discussed previously, as the rain of organic material falls through the water column, CO2 is generated by decomposition of the organics, and because of the increase in pressure, it goes into solution. Thus, tracking a given hypothetical parcel of water through downwelling and subsequent upwelling, the water coming back up to the surface will be enriched in CO2 over what it had when it started its downward journey. What seems to be missing from your analysis is the role of biology in altering CO2 both at the surface and at depths.
“In all cases there is no increase of CO2 in water in the whole trajectory, compared to before the warming, only a decrease everywhere caused by the higher temperatures.”
Exactly! So, less downwells. But, the amount of CO2 upwelling is the same, so it must collect in the surface system.
“But as release and absorption is directly proportional to the pCO2 difference between atmosphere and ocean surface, the original in/out fluxes are fully restored with an increase of 16 ppmv/°C for any ocean surface warming.”
This is an entirely separate process. It is the process of equilibration between the oceans and the atmosphere. It does not force CO2 into the waters. It forces CO2 from the waters into the atmosphere, further depleting the amount that will downwell.
You have not thought this through, Ferdinand. Try to think of it more simply, without the atmosphere involved – it really is the flea on the elephants back. Think of a pipeline, a very long and wide one, containing waters being pumped into one end, and drained in the other. The flow is in steady state.
The source end is warmer than the drain end. We heat up the pipe uniformly along the length. The change in CO2 solubility at the cold end is more pronounced than the change at the warm end, as that is the way temperature sensitivity for Henry’s law works out: for a given delta-temperature, a colder pool of water will outgas more than a warmer pool.
Thus, the distribution of CO2 in the pipeline is altered, with the center of mass shifting to the warmer side. As the pipeline is very long and voluminous, this shift takes a long time. Over relatively short timelines, the content on the warmer side appears to be increasing with the integral of the temperature change.
Just so, with the THC and a warming globe, you have the center of mass of CO2 in the surface oceans shifting towards the equator over a very long time. As CO2 preferentially outgases in the warmer climes, atmospheric content appears to increase with the integral of temperature anomaly. The rise in atmospheric content is merely a symptom of the shifting distribution of oceanic CO2.
Bart:
Exactly! So, less downwells. But, the amount of CO2 upwelling is the same, so it must collect in the surface system.
Bart, a little more careful reading of what I wrote would help: The amount of CO2 upwelling is the same, but a higher temperature at the upwelling and warm areas releases more CO2 into the atmosphere, thus some extra CO2 collects in the atmosphere, while in the ocean surface less remains. That is partly replaced when the waters reach colder areas, but that replacement is less than before if the waters in the cold areas have warmed too. Thus everywhere in the THC / Gulf Stream the residual CO2 level decreases with higher seawater temperatures. There is zero, none, nada build up of CO2 in the surface waters anywhere in warming ocean waters.
This is so elementary that I don’t know anymore what to think of your zeal to defend your theory with such a complete impossible fantasy against all physical principles.
This is an entirely separate process. It is the process of equilibration between the oceans and the atmosphere. It does not force CO2 into the waters. It forces CO2 from the waters into the atmosphere, further depleting the amount that will downwell.
The exchange of CO2 between the oceans and atmosphere is the only process (besides bio-life) that substantially changes the CO2 levels in the ocean surface. The pCO2 difference is the only driving force, which at the upwelling/hot areas is from water into the atmosphere and at the cold/sink side from atmosphere into the ocean surface. The average measured pCO2 difference over a full seasonal cycle is 7 μatm more pressure in the atmosphere, thus the average flux is more sink than source. See Feely e.a.:
http://www.pmel.noaa.gov/pubs/outstand/feel2331/mean.shtml
The change in CO2 solubility at the cold end is more pronounced than the change at the warm end
Your example is not clear at all: while the solubility of CO2 changes with temperature, if that happens in a closed pipe under pressure, that will not give any release of CO2 and the CO2 content remains exactly the same, whatever the temperature, over the full length of the pipe. Thus, the distribution of CO2 in the pipeline is not altered at all.
If the pipe is not under pressure, there may form some CO2 bubbles which in general simply follow the water flow, as long as the bubbles are small enough. Still, no matter the temperature, the mass of CO2 in any parcel of water, including the bubbles, remains the same. Until the bubbles are large enough to collect and separate from the water and then you can have any CO2 distribution you (dis)like…
If that happens in an open drain where you start with hot, CO2-rich water, a lot will be released into the atmosphere at the hot side, less and less as the water cools down and if cold enough it starts to take CO2 out of the atmosphere.
The high concentrations are at the start and rapidly sinks to a minimum and then slightly increases again.
If the water input at the start is heated up and the higher temperature propagates over the whole length, then the initial CO2 release will be higher and over a longer distance before equilibrium is reached and the waters start to take CO2 in again.
The concentrations at the start will go down faster and over a longer trajectory before the slight increase start and thus over the full length the CO2 concentration is lower than in the first case.
That is exactly what happens in an open system which the oceans are: higher temperatures, lower CO2 content everywhere in the waters, but higher CO2 content in the atmosphere and all exchanges are via the atmosphere…
“There is zero, none, nada build up of CO2 in the surface waters anywhere in warming ocean waters.”
This is just silly. You have hypothesized a surface system which has a consistent inflow of additional CO2 from upwelling waters, a decreasing outflow, and yet remains static in terms of overall CO2.
Unless you’ve got a Star Trek transporter device somewhere teleporting the CO2 away to parts unknown, this simply isn’t possible.
“The exchange of CO2 between the oceans and atmosphere is the only process (besides bio-life) that substantially changes the CO2 levels in the ocean surface.”
Not even a starting position. Absolutely ridiculous. Were that the case, The Earth’s CO2 balance would never have changed, ever.
“…if that happens in a closed pipe under pressure, that will not give any release of CO2 and the CO2 content remains exactly the same.”
Nonsense, again. You have a flow in, and a flow out. If these do not balance, the concentration will change.
This is becoming very tedious. I’ve tried every way I can think of to make you open yourself up to the alternatives of your dogma. It appears I need to come up with some other approach to break down your wall. Not likely to happen on this exchange. Until we meet again…
PS: Look over Clyde’s very thought provoking inputs above. I will be incorporating his thoughts into my own outlook, and you will see them again.
Clyde,
In your exchanges with Bart, it strikes me that you have simplified things a little too much. Strictly speaking, Henry’s Law only applies to relatively inert gases like oxygen and nitrogen.
I do simplify things, because it gets already complicated enough, without violating the essence. In the case of CO2, Henry’s law still applies, but only for CO2 as gas, not for bicarbonates and carbonates, That makes that more CO2 is taken away or released by/from seawater that of fresh water. That is about a factor 10, the Revelle/buffer factor is an indication for that. A doubling of CO2 in the atmosphere doubles free CO2 in seawater, as good as in fresh water, but then it stops in fresh water (99% free CO2) but goes on in seawater (1% gets 2% free CO2) and ultimately increases total carbon with about 10%.
The temperature dependency of the CO2 equilibrium between seawater and the atmosphere was established in the literature between 4-17 μatm/°C, field measurements of over 3 million seawater samples are corrected for temperature differences between in situ at the ship’s inlet and the measurement devise at 16 μatm/°C. That is also the average change in CO2 following long-term temperature changes over glacial and interglacial periods.
As we have discussed previously, as the rain of organic material falls through the water column, CO2 is generated by decomposition of the organics, and because of the increase in pressure, it goes into solution.
The discussion with Bart is about the temperature influence, not bio-life. Indeed bio-life depletes the ocean surface everywhere it takes place and a lot of it will come back over time, but the return is an extremely slow process which take hundreds of years and is largely buffered. Temperature changes have a rapid effect on CO2 levels in the surface layer over the full trajectory of the THC and hence in the atmosphere. But there is no way that CO2 accumulates anywhere in the surface layer with higher seawater temperatures…
Bart:
This is just silly. You have hypothesized a surface system which has a consistent inflow of additional CO2 from upwelling waters, a decreasing outflow, and yet remains static in terms of overall CO2.
Bart, if the waters flow from the upwelling zones to the sink zones through a closed pipe, all the supplied CO2 in the upwelling would sink into the downwelling.
Any change of CO2 in the water flow is via the atmosphere, not directly in the water flow itself. Any accumulation of CO2 from warming oceans is in the atmosphere, the CO2 content in the waters from source to sink only gets more depleted everywhere the temperature increases.
“…if that happens in a closed pipe under pressure, that will not give any release of CO2 and the CO2 content remains exactly the same.”
Nonsense, again. You have a flow in, and a flow out. If these do not balance, the concentration will change.
Bart, this only shows that you have no clue about what you are talking: if you have a closed pipe, then what goes in goes out. That is the case for water as good as for its CO2 content. No CO2 can disappear if there is no open contact with the atmosphere. Temperature does change the solubility by changing the pCO2 of CO2 in the water stream, but if there is no air in the pipe and sufficient pressure, CO2 simply remains in the water and flows unchanged in concentration from input to output…
Bart:
You say to Ferdinand;
Sorry I messed up the formatting, but the links work.
Richard
Richard:
The most likely cause of such a change to the equilibrium is the rise in global temperature over recent centuries which is recovery from the Little Ice Age.
The solubility of CO2 in seawater changes with ~16 μatm/°C, which is about 16 ppmv/°C in (dynamic) equilibrium. That is confirmed by over 3 million seawater samples all over the oceans.
If we may assume that the warming since the LIA is not more that 1°C, that means that maximum 16 ppmv of the increase is caused by warming oceans. Vegetation on the other hand grows faster with higher temperatures in the extra tropics, where most NH land is…
The rest of the 110 ppmv increase (of which 80 ppmv since Mauna Loa started) is from human emissions.
Bart,
I wrote:
Bart, if the waters flow from the upwelling zones to the sink zones through a closed pipe, all the supplied CO2 in the upwelling would sink into the downwelling.
Which is true for a fully filled pipe.
In the case that the pipe is half full everywhere, there is a different situation, but not too difficult to follow:
At the upwelling/hot parts, temperature is high, pCO2 of seawater (much) higher than of the atmosphere. CO2 is released into the above atmosphere and waters in that part of the pipe loose some of its CO2.
That gets on for every part of the pipe where the pCO2 is higher than in the air above it and in ratio to the difference.
For the moment we assume that the air follows the water flow at the same speed in the pipe.
Once the temperature is low enough, the pCO2 of the water is less than of the atmosphere above it and the water starts to take up CO2 out of the air. Again in ratio to the pCO2 difference.
If everything is in steady state, water as well as the air above it are at the same CO2 level at the output as at the input, no change in the atmosphere and CO2 sinks = CO2 sources.
Now we heat the pipe over the full length.
That means more release of CO2 in the upwelling/hot parts, thus more CO2 in the above air and less residual CO2 in the water. That remains so for a longer length of the pipe, before pCO2 of water and air are again equal. Because the “cooler” parts also warmed, the pCO2 difference is less and thus less CO2 is taken away out of the atmosphere and the waters are increasing less in CO2 content. Ultimately there is less CO2 remaining in the water at the outlet than at the inlet.
Thus there is a disequilibrium between sources and sinks and somewhere CO2 accumulates. The water now has less CO2 content over the full length of the pipe. All accumulation is in the air above it…
But this is not a physical representation of the system, The air is not contained, and is overturned continuously, as well as surface waters are exchanged with subsurface waters.
I would be surprised to see much of change in pCO2 of general surface water in or out of the tropics, and I would be surprised to see that the sinking polar water did not have a measurable increase in pCO2 as compared to general incoming water heading into the polar areas.
But this would make a great experiment in an Olympic sized enclosed swimming pool.
micro6500,
I fully agree that a pipe, filled or not, does not resemble the water/CO2 flux between upwelling near the equator and downwelling near the poles… A drain open to the atmosphere would be a better example.
For some reason, which I still not understand because physically impossible, Bart thinks that somewhere in the ocean surface, CO2 piles up when the ocean temperature increases, while all what happens is that everywhere the CO2 (DIC) level decreases anywhere the ocean temperature increases,… The increase is in the atmosphere, if not reversed by the increase in the atmosphere by humans, volcanoes or anything else…
Feely e.a. has made an introduction for his compilation of a lot of measurements in the oceans, alluding to the concentration difference between polar and equatorial waters
http://www.pmel.noaa.gov/pubs/outstand/feel2331/exchange.shtml
The maps show in which areas there is a net release of CO2 from oceans to atmosphere and vv., where there is a loss of CO2 (DIC) in the ocean surface and vv. February / August:
http://www.pmel.noaa.gov/pubs/outstand/feel2331/maps.shtml
and next section for the yearly averages…
Ferdinand,
Thank you for the detailed critique. I will need some time to digest your points. It isn’t clear to me how your CO2 Trends graph makes the point that the source of the extra CO2 has to be coming from the NH. I’m glad to see that someone has looked at the role of anthropogenic water vapor. Can you provide me with a citation for that?
Clyde,
The water cycle contribution was a back-of-the-envelope calculation of myself:
The amount of water from burning fossil fuels is roughly 20 Gt/year, waste heat via cooling towers (also from nuclear plants) and rivers, spraying water for irrigation may add the same amount directly or indirectly into the atmosphere or say 50 GtC/year to take it broad enough.
The water cycle is about 500,000 km3 per year (as precipitation according to Wiki), where 1 km3 = 1 Gt. That makes that the human contribution is about 0.01% in average level increase. Even with a 10-fold underestimation, it is hardly of importance. Except maybe if you look at local influences like irrigation in shielded valleys as Dr. Spencer did a few years ago: there was an increase of night time temperatures.
It looks like the fate of the small influence of heat from fossil fuel burning on the earth’s temperature: simply negligible…
——
The trends show that the increase of CO2, no matter the source, is in the NH at ground level first and then takes some time to reach higher altitudes and lower latitudes… That proves that the source of the increase is in the NH at ground level. Further, the decrease in δ13C also is in the NH first:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/d13c_trends.jpg
That proves that the source of extra CO2 is in the NH and has a low 13C/12C ratio. That can’t be the oceans (as these are a lot higher in δ13C than the atmosphere). Neither the biosphere, as that is a net sink for CO2 and preferentially 12CO2, thus relative enriching the atmosphere in 13CO2…
Ferdinand,
I’ll take at face value your estimate of 20Gt/y of water derived from fossil fuel burning. That is pretty easy to check. However, I’m not comfortable with the hand waving estimate of another 50Gt/y from cooling towers and pivotal irrigation. Unless one does all the hard work of determining the surface acreage and draw down of all the major dam projects from Hoover, to Aswan, and more recently in China, then it is hardly even a SWAG. One also has to take into account the aquifer water withdrawn around the world. You neglected industrial activities using water to cool such as the hot steel rolling I mentioned, swamp coolers in arid parts of the world, and water evaporated in the cooking of food for 7 billion people. In summary, a complete analysis of the contribution of anthropogenic water vapor has not been done. Pardon me if I don’t accept your claim.
I agree that the waste heat of energy consumption is negligible.
Your graph does not speak to isotope ratios, despite you raising the issue in your response. I’d like to comment on the Point Barrow data. When I was last there, all the homes were heated by natural gas from a nearby gas field, which probably flared excess gas. The locals burned all their trash in open 55-gal oil drums. I don’t know what the locals did for sanitation, but at the army base, the toilets were essentially blowtorches in a bowl. One never sat on a cold toilet seat. The feces were incinerated to a small pile of ash, releasing CO2 (and water vapor). Almost everyone owned at least one snow mobile, which they used even when there wasn’t snow on the ground. Airplanes landed nearby regularly. I believe all the electricity came from diesel generators. The Army and some civilians had trucks and cars. I wouldn’t put much stock in the CO2 levels reported from Barrow. I don’t follow your explanation with respect to altitude. Mauna Loa is the highest recording station, yet has the highest values if we exclude the sea level station at Pt. Barrow.
Clyde,
Barrow only uses CO2 values measured with wind from the seaside for daily, monthly and yearly averages, the stations is situated on a land tongue somewhat more in the sea. Other stations like Alert (ALT, Nunavut, Canada) show the same trend as Barrow. Both the seasonal variability and the increase in the atmosphere are more and earlier visible in low lying stations than at altitude. It takes a few weeks to a few months to disperse the changes with altitude and latitude… Here for the seasonal changes:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/seasonal_height.jpg
After some more digging: of the 500,000 Gt/year water cycle, already 400,000 Gt/year falls in the oceans, according to Wiki. 100,000 Gt/year is what falls on land.
Form another source:
Only ~1% of the world’s fresh water is accessible for direct human use
If we may assume that – until now – all fresh water reservoirs are replenished by rain, humans have access to about 1% of 100,000 Gt/year water or 1,000 Gt/year.
Not all water used by humans evaporates. Even as 70% of all water used by humans is for agriculture, a lot of that still is absorbed by plants and soil and part of the latter flows back to rivers. The same for households and most of what is used by industry (except for cooling towers).
Even if all 1.000 Gt was explored by humans and all that water evaporates, that still is only 0.2% of the full cycle, but in reality it is a lot lower…
Ferdinand,
So, it sounds like Barrow has to go without readings occasionally and there is a further assumption that the winds are always linear and not cyclonic.
There are some caveats in yous statistics about fresh water availability. I suspect that the reason the ‘available’ fresh water is so low is because they are counting the water locked up in ice in Greenland and Antarctica.
One of the real problems confronting agriculture is that, in most places in the world, aquifers are being depleted at a rate much greater than their re-charge rate. Some have suggested that this is contributing to sea level rise because whether it runs off or precipitates, most of it ends up in the oceans.
Water that is taken up by plants either leaves by transpiration or is released when the plants are consumed.
I think that something that you are overlooking was that my analysis was based on land temperatures and how humanity might be impacting land temperature directly. So, what needs to be examined is the hydrologic cycle focusing on the land. I think that when analyzing this issue, the oceans and land have to be examined separately, but in parallel.
Clyde,
Barrow notices all the (hourly averaged) raw data plus the stdev over that hour. They only don’t use the data for averaging over days, etc. if the wind is from land side or the data are highly variable within that hour (contaminated)… Here the raw data for 2007-2008:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/co2_trends_brw_2007_2008.jpg
The water cycle indeed needs more work, but I still think that the human contribution to that cycle is globally negligible, locally/regionally it may be of importance.
For the CO2 cycle, higher temperatures in general give more CO2 uptake by plants and more release by the oceans, where on long term the oceans wins the battle… But that is quite limited by Henry’s law. The opposite effect from CO2 on temperatures is even more limited…
ferdinand , regarding ocean sinks of co2 ,are you accounting for plankton ? some species vary by orders of magnitude regionally over long and short time scales . i have no idea of the total effect as a result, but looking at the literature neither does anyone else.
bit chilly,
There are few investigations about the impact of plankton on CO2 levels in the oceans and atmosphere, only an overall inventory of the total biosphere from the oxygen and δ13C balances. As the oxygen content of the ocean surface in general is saturated, any uptake of CO2 by plankton will release oxygen and that will get into the atmosphere. Reverse for decaying and eaten plankton in the food chain of the oceans. Taking into account the solubility of oxygen in seawater with temperature, one can calculate the total change from land + sea biosphere. That gives that some 1 GtC/year as CO2 is absorbed by the total biosphere…
Specific for the Bermuda area, there was an investigation over several years to see what the impact was of bio-life on CO2/bi/carbonate and pCO2 levels in the ocean surface waters:
http://science.sciencemag.org/content/298/5602/2374.abstract
The full article is freely available after registration.
Net effect seems to be a +/- 50% variability around the average sink rate, which is a matter of wind mixing with the deeper ocean waters, which bring more nutrients to the surface and hence more bio-life…
Ferdinand,
At least one of the OCO-2 maps strongly suggests that a plankton bloom in the North Atlantic has noticeably reduced the atmospheric CO2.
Ferdinand,
It just occurred to me that there is yet another problem with your interpretation of the CO2 trend graph above. As I understand your position. diffusion of anthropogenic CO2 lags at Mauna Loa because it is some ~ 3400m higher than Barrow. However, I’m pretty sure what is being measured is the result of orographic uplift and was at sea level a few tens of minutes before being recorded. Also, since these are volume readings, have they been converted to standard temperature and pressure so that they can actually be compared?
Clyde,
What is measured at Mauna Loa is mostly from the trade winds, which blow in from Africa over thousands of km over the oceans. Sometimes they have downwind influence from the volcanic vents (+4 ppmv, highly variable) and in the afternoon upwind conditions (-4 ppmv, depleted by vegetation in the valleys). But indeed the differences in Hawaii between sea-level (Kumukahi) and 3,400 m height (MLO) are very small:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/co2_trends_mlo_kum.jpg
More a matter of dispersion from the first few hundred meters over land into the bulk of the atmosphere over all latitudes than of altitude I suppose…
Ferdinand,
Here is a link to an OCO-2 animation for the first year: https://www.youtube.com/watch?v=_UEZqyGU5RU&feature=youtu.be
If you haven’t seen it, you should find it interesting. There is little, if any, compelling evidence for anthropogenic sources when looked at in the big picture. It seems that the major source of NH CO2 is probably decomposition of organic material, once it starts to warm up; that is then removed when the trees leaf out. I don’t see the evidence of the Ferrel cells you said were providing the CO2 to Barrow.
Clyde,
The only partly month (05/12 to 05/29) where Barrow is covered shows the same red (402-405) from the mid-NH latitudes to the high North, which would be remarkable as there is only tundra, no big trees on the North Slope. Thus far les debris to decay than in the mid-latitude forests of Canada and Siberia. The main possibility is that the extra CO2 was transported from the mid latitudes.
The same for the “CO2-hotspot” between the coast and Greenland, which is probably blown in from the NE USA and the hot-spot over West-Africa, where Nigeria is a known huge emitter by flaring (legal and illegal) of lots of waste gas. etc, but the red blob is far more widespread.
I have no idea if and how they take into account the spread out of the sources by wind…
Ferdinand,
I’m familiar with Barrow, having spent a month there. As I understand the operation of OCO-2, they don’t take into account wind. These are supposed to be actual, calibrated values as observed from orbit. That’s why I though you would find it interesting. I tend to give precedence to actual measurements over theory.
There is one human activity where water vapor might have a significant effect over time in all latitudes, that does not ‘induce’ vapor, but triggers nucleation and condensation: jet contrails and their seeding of clouds where clouds would not otherwise have been… during borderline (but stable) conditions, or in absence of other natural factors (eg. cosmic rays). The sight of contrails over the desert surrounded by clear sky or spectacularily visible from space empirically suggest this may be an important factor to consider for the latter part of the industrial era.
I see that the presence of clouds is observed to decrease the Diurnal Temperature Range by 25-50%, and in the days following 11-Sep-2001 when aircraft were grounded David J. Travis noticed a 1.8C rise in DTR,
What of this ‘anthropogenic’ effect, how would one factor for it, and and how might air travel since the late 20th century affect these global min/max? Thx.
“Really? Does BEST actually claim that there is no such thing as the UHI? Or is their claim more subtle than that?”
1. UHI is real
2. You can find UHI at INDIVIDUAL SITES
3. you can find smaller effects in REGIONS
ftp://ftp.ncdc.noaa.gov/pub/data/ushcn/papers/hausfather-etal2013.pdf
4. When you try to find UHI in the GLOBAL RECORD…… It gest lost in the noise.
So, UHI is real. If you pick an individual site and look at CLEAR WINDLESS DAYS.. you will
see big values. Thankfully some, but not all, of these can be corrected by homogenization.
Take Seoul
http://berkeleyearth.lbl.gov/stations/156456
The PRECIOUS RAW DATA… shows warming of 2.3C per century
and QC and breakpoint…. Warming is 1.4C per century
or take this city
http://berkeleyearth.lbl.gov/stations/173631
Or take RENO.. Anthony has measured UHI there.
Here is what homogenization does
http://berkeleyearth.lbl.gov/stations/173102
Reducing the warming from 1.85C per century to .7C per century
OMG.. we are adjusting stations DOWNWARD!
Bottomline: UHI is real. It exists. However, finding that Signal in GLOBAL DATA is extremely difficult
because
1. AVERAGE monthly UHI is relatively small even in large cities
2. Some “urban” stations show NEGATIVE UHI— basically from micro site
3. A quantitaive definitive definition of “urban” is always subject to uncertainty
4. The urban landscape is VARIED.. some areas have negative UHI, others strongly positive
5. Some rural landscape are naturally warmer than Urban areas, especially those in areas of
little water, or areas where the land form conspires to make the rural more warm than the urban
Finding ANY signal in global data is extremely difficult, cough, AGW….
CO2 curve also correlates well with my body weight.
“””””….. Thankfully some, but not all, of these can be corrected by homogenization. …..”””””
Why would you ” correct ” anything that IS a part of the real Temperature record ??
G
Las Vegas is toast. Seriously. Thanks to the urban heat island effect, Sin City is 7.3 degrees hotter on average than the surrounding hinterlands.
having read this thread in its entirety http://rankexploits.com/musings/2016/richardlh-is-now-subject-to-willardian-moderation/
i would recommend it for anyone that wants to understand what best actually claim vs what some people think they are claiming. it would save a lot of arguing at cross purposes in the future on all climate blogs if everyone read it.
Forrest,
No, BEST does not claim that UHI doesn’t exist, nor did I. Perhaps I could have stated it differently. I suspect that plumes of heat, moisture, and particulate pollution are influencing ‘rural’ stations downwind of cities.
If the UHI effect is corrupting the result; that by itself is prima facie evidence that the data set sampling regimen is invalid.
There is no negative connotation to urban heat islands. if they heat faster during the day, and reach higher Temperatures, then they also are radiating at a higher than average rate, and improving the night time cooling rate.
But if you insist on using an UHI temperature as a sample for some place 1200 km away, then no wonder you get screwed up results.
The first lecture in climatology should be on the general theory of sampled data systems.
G
The first thing that Clyde should have done was look at the residuals of his regressions. If he had done so, and understood what he was looking at, this would never have been written.
Basil
Basil,
OK, if you are going to be cryptic, I guess I will actually have to do what you have suggested. Can you at least give me a hint of what I’m looking for?
Clyde,
As I understand it, you are regressing BEST temperature data against the log of CO2 PPM. Before doing that, look at the two series over time:
http://i.imgur.com/FojmOx3.jpg
This is anomaly data, not temperatures, as that is what was in the BEST file you linked to. But that shouldn’t matter. If you are regressing BEST data on the log of CO2 PPM, you end up with a pattern of residuals over time that looks like this:
http://i.imgur.com/W4umWHy.jpg
There should be no evident trend in this graph. The residuals should be randomly distributed so that there is no trend over time. Looking back at the first graph, you can see that what you are doing is effectively rotating the BEST data counter clockwise around the CO2 data to force a fit that does not exist in the real world, systematically underestimating the relationship in the earlier years and overestimating it in later years.
Analysis of time series requires some careful diagnostic skills and looking at residuals is basic. Also, a statistic like the Durbin-Watson statistic implies a problem with a simple OLS regression of these data. It should be around 2.0, and for the regression I ran it was about 1.2, indicating serious serial correlation over time. Time series analysis usually requires alternatives to OLS. I cut my teeth on Cochrane-Orcutt (which will tell some how old I am). But it gives the following pattern of residuals over time:
http://i.imgur.com/7QoAPYt.jpg
Notice there is no trend in the residuals over time. And the Durbin-Watson of the resulting regression is 2.15.
Hope this helps.
Basil
Basil,
Your first assumption is incorrect. I converted the anomalies to actual temperatures, as shown rather clearly on the y-axis.
Clyde,
From the file you linked to, how did you covert anomaly to temperature?
Basil
Basil,
The metadata at the beginning of the file describes how BEST derived their anomalies from the original temperature data. I added back what they subtracted. I feel that anomalies often obscure information.
Clyde, your missing two steps, first is to determine the physical relationship (blcjr has done this by plotting the two on top of each other and noticing there isn’t any obvious relationship). plot the two series on the same timeline and compare what you have. Then if appropriate run a scatterplot. (not appropriate in this case the two are obviously incoherent). After the scatterplot, look at your residual (blcjr has done this as well and you don’t have a random cloud so the relationship is not linear). If the linear regression is meaningful you should get a residual plot that is random. If there is a pattern then your linear regression is not valid, meaning there is no linear relationship. This is basic statistics taught in entry level college statistics classes. If your going to use statistics you should at least learn the rules of use and the many tests you can use to validate your work.
v/r,
David Riser
David,
Thank you for your patronizing remarks. I’m running as fast as I can and I don’t have many years left. I had basic statistics in college in both the psychology department and the mathematics department. My MS was a statistical analysis of heavy-mineral concentrates as they correlated with platinum group minerals. Yet, somehow, I seem to have missed your basic advice along the way. Perhaps I have just forgotten it because I have jammed so much into my skull over the decades. Incidentally, you seem to have missed that Basil made an erroneous assumption about using anomalies. Does that influence your advice in any way?
David,
BTW, the correct spelling is “you’re.” They teach that in basic English. If you are going to write criticisms, you should learn the rules of use.
Clyde,
I guess I am not sure what you’ve done with that metadata. Once adding the monthly averages back to the anomaly, you would have two data series like this:
http://i.imgur.com/Zmrb1PA.jpg
But you seem to have selected just two subsets of the data, something called “high temps” and another called “low temps.” I don’t see where you explained how you did that, but I would have to surmise that you are using just selected months out of the year, and not all the data. Could you explain that to me?
Basil
Basil,
For a start, I’d suggest going back to a previous post I made: http://wattsupwiththat.com/2015/08/11/an-analysis-of-best-data-for-the-question-is-earth-warming-or-cooling/
I am using all 12 months of data for each and all years in the stated time series. Basically, I added the BASE-PERIOD AVERAGES to the monthly anomalies for both the BEST high-temperature series and low-temperatures series. Those averages are in the metadata that precedes the actual tabular data.
Clyde, Thanks….In all that time and work you never performed a residual plot to determine if your regression was legitimate? Or to refine your model. You never took a look at the physical relationship to see if what your doing makes sense. I think you should pull your old stats books and refamiliarlize yourself with science of statistics.
v/r,
David Riser
David,
You are still misspelling “your”. “I think you should pull your old” English “books and refamiliarlize [sic] yourself” with English grammar.
Clyde,
Your own analysis demonstrates that your regression is meaningless. If there is a relationship, its not going to disappear as you go back into time. Temperature and CO2 are related to a great number of variables so the likelihood that one is dependent to the other is very small. Using the averages of a highly adjusted surface temp to the averages of a single measurement of the globe for global CO2 and achieving any kind of correlation is related to luck. In short, your abusing statistics for the internets amusement.
v/r,
David Riser
clyde ,worthj a read. http://www.wikihow.com/Use-You're-and-Your
Shouldn’t you have sent this to the attention of David? I’m not the one having problems with the words.
Ferdinand,
You convinced me a long time ago that the rise in CO2 can be attributed to human emissions. But the real issue is this: has the rise in CO2 caused any global harm, or damage of any kind?
We know for a fact that the rise in CO2 has been beneficial to the biosphere. Agricultural productivity is rising as a direct result of the added CO2. On any cost/benefit analysis, more CO2 (at both current and projected concentrations) is a net benefit; more is better.
Based on all available evidence, at current CO2 concentrations Earth’s climate is insensitive to more atmospheric CO2. The predicted ramp up of global warming never happened. It was a false alarm.
Ferdinand, you said pretty much the same thing here a while back:
“In my opinion even a doubling (of CO2) would have little impact, as clouds are a negative feedback (while all current GCM’s include clouds as a positive feedback!), thus a doubling of CO2 would have only moderate (and thus globally positive) effects.”
So the rise in CO2 is a net benefit. That negates the ‘carbon’ alarmism, no?
dbstealey,
I still stand completely behind those wise words of an old man… (just kidding).
Indeed I haven’t seen any harm from the increased CO2 levels and don’t expect any harm from slightly elevated temperatures, and even more benefits from increased crop yields thanks to more CO2…
All the alarmism is based on completely failed climate models…
Ferdinand,
Thank you for that. Those were indeed wise words, and they completely negate the basic premise of the climate alarmists.
The basic debate has always been over the claim that rising CO2 will cause runaway global warming and climate catastrophe. They deny it now, but that is because Planet Earth is busy falsifying their failed hypothesis: as we know, global warming stopped many years ago.
But without their basic (false) alarm — the claim that CO2 emissions will result in very bad consequences — what have they got?
They’ve got nothing. Their conjecture has been falsified. At this point the proper thing to do is to reject their CO2=cAGW conjecture, and try to formulate a new one that is able to make accurate predictions (since correct and repeated predictions are the required standard of a valid conjecture, hypothesis, theory, or law).
But instead, they dig in their heels and try to defend their original hypothesis, because they think that if they admitted they were wrong originally, that would mean the hated skeptics were right all along (which is true).
To defend their original false alarm, now they are claiming that global warming never really stopped at all. And a related claim: that satellite data is no good compared with adjusted surface station ‘data’.
“If an honest man is wrong, after demonstrating that he is wrong, he either stops being wrong, or he stops being honest.”
The climate alarmist crowd stopped being honest a long time ago.
If I read fig 1 correctly, it suggests we have seen almost 2C of temperature rise between 1958 and 2014. From 14C to almost 16C. That is much more than I have seen suggested elsewhere. I suspect if claims of a 2C rise in temperature in the last 50 years could be substantiated it would be shouted from the rooftops as categorical proof of CAGW. Indeed I remember well a recent article claiming most emphatically that the world had seen close to 1C rise since 1900. That’s a long way from nearly 2C rise since 1958.
I question plotting monthly values since there is a strong seasonal cycle in Mauna Loa CO2 data (and also in the temperature data related to the fact that the northern hemisphere has a higher proportion of land area which heats and cools faster than the oceans with the seasons). At the least I think it would be necessary to remove this trend via a 12 month boxcar filter on all the data.
Further, according to the Mauna Loa data the CO2 conc in 2000 was 370 ppm while in 2014 it was about 400 ppm. That is a log change from 8.53 to 8.64 or an increase in Log(CO2) of 0.11. Using the trend line equation from fig 1 that should give an increase in temperature of 3.626*0.11 = 0.4C yet the satellite record shows no warming over this period (the pause) and even the most extreme land/ocean surface record shows only about 0.1C rise. Something does not seem to tie up. The temperature rise over the entire satellite era is only about 0.4C.
Plotting data in this way only makes sense if there is no time constant between cause and effect. ie: the impact of rising CO2 on temperature is more or less instantaneous but the CAGW fraternity continuously tell us that there is a long time constant involved and even if we stabilise CO2 levels now temperature will continue to rise for decades. If that is the case the plots in the article don’t really mean anything because the same CO2 concentration at two different times would give quite different temperature outcomes depending on the past history.
Looking at the Mauna Loa data it is clear that CO2 concentration since 1958 is almost linearly related to time. Thus plotting temperature versus CO2 becomes essentially the same as plotting temperature versus time. So is the relationship between temperature and CO2 or temperature and time. Indeed, many things change more or less linearly with time. If I relabel the X axis by time (or indeed anything else that changes in a close to linear fashion with time) instead of CO2 does it mean I have proven a relationship between temperature and the new X parameter? I remember someone doing exactly that using the price of postage stamps as the X axis parameter – global temperature linked to the price of postage stamps!!!! More significantly, there is a suggestion of underlying solar activity changing close to linearly between 1958 and about 2000.
I apologise for sounding critical but it seems to me this analysis has some very serious flaws.
Michael,
The problem is that what invariably gets reported in the media and even most scientific reports is the average temperature; I’m showing the high and low temperatures, which are behaving differently. The Scripps CO2 data have been smoothed. Yes, I understand that the satellite data does not show the same warming the the land surface temperature data does. That is an entirely different issue. Any monotonic function increasing over time will show a correlation with time because the function is correlated over time. The question that has to be asked, does the relationship make sense, or is it obviously trivial?
This is a very relevant point to make in view of the recent corruption of the SST record by Karl et al by using temporally biased night time marine temperatures to “correct” fully daily SST temperature records.
Most of the Scripps series are now available at daily resolution.
Speaking of the OCO-2 matter, it seems to me not enough people here are homing in right there, and ask why heavy co2 appears to be coming from the warm Pacific ocean and tropic forests. Thinking along the lines of the diurnal co2 intake/outflow cycle of chlorophyll based microbes due to the metabolism/respiration using co2 dissolved in the water during uptake but co2 is released in tiny bubbles that float to the surface and into the atmosphere during respiration, or something along those lines.
Makes me question whether the increase in atmospheric co2 being evident might be merely from the greening of the Earth since the colder period a couple of centuries ago and not necessarily from trees and plants above the surface. The tropical forest always have plenty of pools of water near the equator.
Anyone know more specifics on the botany/biology along these lines explaining the excess of dissolved chlorophyll laden microbes handling of the co2 flux between the liquid water surface layer and the air above? Could this be much of the net increase out of the control by mankind?
wayne,
There is a way to determine what the contribution is of the biosphere (land en water plants, bacteria, molds, insects, animals) to the CO2 level in the atmosphere: the oxygen balance.
Fossil fuel burning uses oxygen. That can be calculated from sales and burning efficiency. Plant growth on land and in water uses CO2 and releases oxygen. Plant destruction by the rest of the biosphere as feed/food uses oxygen and releases CO2. If you make the total balance, the biosphere is a net source of O2, thus a net sink for CO2 of ~1 GtC/year. See:
http://www.sciencemag.org/content/287/5462/2467.short
and
http://www.bowdoin.edu/~mbattle/papers_posters_and_talks/BenderGBC2005.pdf
Christopher Keating at 11.35 on the 26th Feb stated “it is no accident that the noise in the CO2 signal is correlated to temperature. Now, besides your correlation graph, what other evidence do you have?”. Um one point, if you integrate this “noise” in CO2 ie: just sum it year over year you get the actual plot of CO2 versus time. If this is simply noise in the CO2 signal then the entire CO2 signal is simply the integral of noise and noise is not correlated to anything. I think it would be more accurate to state that the derivative of CO2 conc ie: the rate of change of CO2 concentration is correlated to temperature and that’s a VERY different proposition. One that actually has a very plausible physical explanation – ocean outgassing with a long time constant – the ocean overturning time is around 800 years after all.
Congradultions Mike!!!!
…
You say, “if you integrate this “noise” in CO2 ie: just sum it year over year you get the actual plot of CO2 versus time”
…
Wow…..who would have known that integration is the inverse operation of the derivative?????? Mmmmmm…..I believe they cover that in 1st year Calculus right?
..
What do you say about Bart’s “correlation” proving causation?
..
If the rise in CO2 is due to ocean out gassing, why don’t you show evidence of this by showing that the southern hemisphere CO2 concentrations are higher than the northern hemisphere concentrations due to the fact that the SH has more ocean than the NH???? Better yet, why not measure the CO2 near ocean upwelling points and compare that to downwelling points?????
I don’t see why Clyde chose to look at land temps for this exercise. SST is the relevant metric for global temperature rise since this represents the bulk of the surface and the bulk of the heat content.
However, if we look at the fully expanded graph rather than falttening it as in fig 1 we see at ‘plateau’. Since CO2 got to about 373 ppmv, There is no rise in temperature with CO2.
Now if we look at rate of change of CO2 at MLO we also see a plateau in CO2 rise. The relationship to hot El Nino years is also readily apparent.
http://climategrog.files.wordpress.com/2015/04/mlo_ddt_co2.png
There does not seem to be a similar plateau in human emissions.
http://climategrog.files.wordpress.com/2014/06/co2-log-rise.png
Mike,
I agree that the oceans contain more energy than the land. However, because it changes temperature more slowly with absorbed energy, I felt that land temperatures would be more sensitive to changes in forcing. Besides, the land is where most of us live.
Christopher I think you misunderstood my point. Your earlier post suggested that that in the plot showing good correlation between temperature and rate of change of CO2 the changes in CO2 were just “noise”. What I was trying to point out was that its not noise its actually the first derivative of the CO2 signal ie: the entire rate of change of CO2 with time (not just a noise component on the top of the signal). That means that the remarkable degree of correlation between the rate of change of CO2 and temperature is indeed significant and justifies asking the question why. Sure correlation does not equal causation but neither does it rule it out. Whether the CO2 level is different in the two hemispheres depends to a great degree on the degree of atmospheric mixing between the two hemispheres. Again expecting a difference between upwelling and down welling points depends on the degree of mixing and on the rate of outgassing. After all, a soft drink which is massively supersaturated wrt CO2 does not go flat the instant its opened. The outgassing takes quite a while.
Mike,
There are several years that the correlation between temperature increase and CO2 increase is negative. The full period 1973-1992 and 1998-current. You are just plotting the noise around the trend by looking at the derivatives.
The trend of the derivative for a linear increase of temperature (and the CO2 release caused by temperature) is essentially zero with a slight offset. The trend of the derivatives from both the CO2 emissions and the increase in the atmosphere is linear positive, because both show a slightly quadratic increase over time.
The year by year uptake of CO2 somewhere in nature is highly variable between 10% and 90% of human emissions. Over 10-20 year periods that is between 40-60%, over the full period 50-55%. The average uptake is highly linear and solely depends of the pCO2 difference between CO2 in the atmosphere and the average pCO2 of the ocean surfaces. The latter is modified by temperature. If one calculates the sink rate based on the pCO2 difference with a net sink rate of 2.15 ppmv/110 ppmv (2012) pCO2 difference over the dynamic equilibrium, then the theoretical increase of CO2 in the atmosphere is a nice fit within the temperature caused noise:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em6.jpg
All that says is that the CO2 increase in the atmosphere is largely caused by human emissions and that temperature modulates the uptake with high variability on short term but very moderate on longer term…
The effect of CO2 on temperature is very modest: theoretically about 1°C for 2xCO2, without any positive (climate models) or negative (real world) feedbacks. The opposite also is very modest: 16 ppmv/°C. Hardly worth any discussion…
“There are several years that the correlation between temperature increase and CO2 increase is negative.”
No. Trend lines on uncertain data do not establish correlation. When you can look with your own eyes, and see that the features all match with astonishingly high fidelity, you know you have a significant correlation.
Bart,
There is a huge correlation between temperature/derivative and the CO2 rate of change. No doubt of that.
The problem for your theory is that variability and trend are from different processes. Thus the high correlation between temperature/derivative and CO2 variability says next to nothing about the cause of the CO2 trend…
“The problem for your theory is that variability and trend are from different processes.”
It is mere assertion.
Come on Bart,
There is no doubt that the variability is the response of vegetation to temperature variability.
There is no doubt that vegetation in recent decades is a net, increasing sink for CO2 – the earth is greening,
But of course you reject anything that may undermine your theory…
“There is no doubt that the variability is the response of vegetation to temperature variability.”
The source of the variability and the trend is the same. This match is no accident:
http://i1136.photobucket.com/albums/n488/Bartemis/temp-CO2-long.jpg_zpsszsfkb5h.png
It does not matter if you think they should be from difference sources. This plot shows they are not.
“There is no doubt that vegetation in recent decades is a net, increasing sink for CO2 – the earth is greening…”
But, vegetation also decays, becoming a source. So, you are on very speculative ground claiming it is a net sink.
“But of course you reject anything that may undermine your theory…”
So do you. The difference is, I reject narratives created to rationalize evidence for which other possible explanations exist, while you reject obvious correlations that are right in front of your eyes.
Ferdinand – the dual match between the trend and the variation in that plot is not going to go away, and dismissing it as mere coincidence is not satisfactory or compelling.
There have been many such surprises in science history where physical evidence upended well-established thought. That is why scientific rigor requires that you provide good reason for disregarding what the evidence shows.
You have not provided any such reason. You have only kept going around in circles insisting that your notions must be true, and therefore one must dismiss this striking congruence as happenstance.
It doesn’t work. It’s not enough. The odds of this phenomenal match occurring by accident are too long.
“Science is not about odds, it’s about data.”
And, the data show an amazingly good match between the rate of change of CO2 and temperature anomaly across the entire spectrum.
BTW, from Occam’s razor to quantum mechanics, science is all about odds. You obviously have no clue about what science is or is not.
Bart,
the dual match between the trend and the variation in that plot is not going to go away, and dismissing it as mere coincidence is not satisfactory or compelling.
The dual match between the trend and the variation is the result of a real match between the transient response of vegetation to temperature variability and the arbitrary factor and offset you used to match the slopes: pure curve fitting, nothing else.
One can match any linear slope with any other linear slope whatever with the right factor and offset. Which doesn’t say anything about its physical and/or chemical correctness.
The response of the biosphere to temperature variability is more decay for increasing temperatures and vv. with a response time of less than a year. That zeroes out in 1-3 years. That is observed. The response of the biosphere to prolonged warming over 3 years and longer is more uptake. That is observed. Different processes at work for the variability and the slope not the slightest shred of doubt about that.
Moreover, combining the transient response of vegetation on temperature variability with the response of the oceans (and vegetation) on the increased pressure in the atmosphere by human emissions gives exactly the same result:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/rss_co2_emiss_deriv.jpg
Which fits all observations, while your pure mathematical match of trends fits none…
That is the problem with all the alternative theories of a non-human cause of the increase in the atmosphere: all fail one of more observations, thus should be rejected from the start on. Yours even fails ALL observations, except for the nice arbitrary curve fitting…
Bart:
But, vegetation also decays, becoming a source. So, you are on very speculative ground claiming it is a net sink.
Bart, for the tenth time I suppose, it is possible to know the balance of CO2 uptake and decay of the biosphere as a whole (land and sea plants at one side, bacteria, molds, insects, feed and food on the other side) by looking at the O2 and δ13C balances.
Biological processes either capture CO2 and preferentially 12CO2 and release oxygen or release CO2 with low 13C and use oxygen. Ocean CO2 releases / uptake doesn’t change oxygen, besides its solubility with temperature and hardly changes δ13C.
Careful measurements of the oxygen and δ13C changes since ~1990 show that the biosphere as a whole is a net, growing source of O2, thus a net sink for CO2 and preferentially 12CO2. The earth is greening.
The latter is confirmed by satellite measurements of chlorophyll all over the globe and especially in semi-arid areas.
Thus nothing speculative, only observations…
“One can match any linear slope with any other linear slope whatever with the right factor and offset. Which doesn’t say anything about its physical and/or chemical correctness.”
It matches both the trend and the variation. You are fooling yourself. You don’t get this kind of agreement by happenstance.
“Moreover, combining the transient response of vegetation on temperature variability with the response of the oceans (and vegetation) on the increased pressure in the atmosphere by human emissions gives exactly the same result.”
Not exactly, and it is very contrived. With 5 parameters, you can fit anything to some degree. You have simply massaged the data into a form you find pleasing. But, the simplest explanation, the one which requires by far the least processing and the fewest assumptions, is that temperatures are driving virtually the entire rate of change. Occam’s razor then comes heavily down on the side that, that is what we are seeing.
“Yours even fails ALL observations…”
No, it just fails to fit your narrative. You are rationalizing a result you prefer, and ignoring data which disconfirms it.
Bart,
It matches both the trend and the variation. You are fooling yourself. You don’t get this kind of agreement by happenstance.
It matches the variability because that is a real cause and effect, It matches the slopes because you used an arbitrary factor and offset, which always can match any two linear slopes, thus has no bearing in any physical process, the more that the slope is proven not caused by the same process as the variability.
You are the one who fools himself…
Not exactly, and it is very contrived. With 5 parameters, you can fit anything to some degree. You have simply massaged the data into a form you find pleasing.
Because you don’t like the result, it must be “manipulated”? The shift of the data from the transient response of CO2 to temperature changes are exactly at the same timing as your “match” with the temperature data. The data of the slope are from the linear response of the increased pressure in the atmosphere by human emissions per Henry’s law for the temperature of the oceans. All simple, direct calculations, no manipulation with arbitrary factors and offsets as in your plot.. Here detailed for the period 1987-2002 with the largest temperature changes:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/rss_co2_emiss_deriv_1987-2002.jpg
the simplest explanation, the one which requires by far the least processing and the fewest assumptions, is that temperatures are driving virtually the entire rate of change.
The simplest explanation must fit all observations, or it is not the right explanation.
Your explanation that temperature is the cause of both the variability and the slope is based on the assumption that both are caused by the same process(es), which is proven false.
Your assumption that the oceans are the main cause is proven false, due to the too high 13C/12C ratio and over 3 million seawater samples which show that the oceans are more sink than source. The same for any piling up of CO2 in the oceans as cause, which is physically impossible.
Your assumption that the natural cycle is dwarfing the human emissions is proven false, as that is already 6% of the natural cycle and twice the natural variability in residuals after a full seasonal cycle. Moreover, the fourfold increase in net sink rate is not caused by a fourfold increase in natural cycle, which would be necessary to dwarf the fourfold increase in human emissions. There is no sign of such an increase in natural circulation.
Further, we have some more items which aren’t matched with your theory:
– the oxygen balance.
– the δ13C balance.
– Henry’s law for the solubility of CO2 in seawater.
– the impossibility of a continuous influx of CO2 from a step increase in temperature without any negative feedback from the increased CO2 pressure in the atmosphere.
Is there in fact anything in the real world which supports your theory besides your artificial match of two linear slopes?
No, it just fails to fit your narrative. You are rationalizing a result you prefer, and ignoring data which disconfirms it.
Have you any data which shows that “my” theory, based on real world processes, doesn’t fit any observation?
“It matches the slopes because you used an arbitrary factor and offset, which always can match any two linear slopes…”
No, Ferdinand, you cannot match the trend with an arbitrary scale and offset. That would cause a mismatch with the variation. It fits so well because both the trend and variation are from the same process, and the scale factor and offset that match both of them are the same.
“Because you don’t like the result, it must be “manipulated”?”
No, because it required so many arbitrary assumptions, and eliminated the trend that was already in agreement between the rate of change of CO2 and the temperature.
“Have you any data which shows that “my” theory, based on real world processes, doesn’t fit any observation?”
Yes. Your theory fails because it arbitrarily eliminates the trend in dCO2/dt and T which already match one another very well.
Bart:
No, Ferdinand, you cannot match the trend with an arbitrary scale and offset. That would cause a mismatch with the variation. It fits so well because both the trend and variation are from the same process, and the scale factor and offset that match both of them are the same.
There is a wide range of possibilities in matching the trends and the variability if you don’t take it too stringent, simply because the slopes are not too different. For over halve the full period, the factor is negative, causing an upside down variability between temperature and CO2 rate of change. Which shows – again – that variability and slope are completely independent of each other.
The transient response of vegetation to the variability of temperature shows exactly the same variability and timing in the derivatives, but has zero slope:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/rss_co2_nat_deriv.jpg
Only the amplitude is overblown, thus needs a smaller conversion factor from temperature to CO2.
Again you reject the observations that the variability is proven from vegetation out of the opposite CO2 and δ13C changes, while the trend is proven not from vegetation. These observations are rock solid. If you have any proof that vegetation is not involved in the variability and/or that vegetation is a net source of CO2 over periods longer than 1-3 years, you may have a point. Until then your “temperature explains everything” is refuted.
No, because it required so many arbitrary assumptions, and eliminated the trend that was already in agreement between the rate of change of CO2 and the temperature.
In “my” theory there are only two assumptions: most of the variability is caused by the influence of temperature (where we agree) on vegetation (where you have no opinion) and most of the increase is caused by the difference between human emissions and the net sink rate, which is proportional to the extra CO2 pressure in the atmosphere above steady state (which you completely ignore) for the momentary temperature. Quite reasonable assumptions…
Further: your assumption is that the trend is caused by temperature, which violates every single observation, thus your remark that the trend is already accounted for has no base in reality and doesn’t disprove another theory, which fits all observations…
Again, when you match the single scale factor for variability, you also match the slope of the trend. One single factor does all of that. That is far too unlikely to be mere happenstance.
“Again you reject the observations that the variability is proven from vegetation out of the opposite CO2 and δ13C changes, while the trend is proven not from vegetation.”
As William Shakespeare’s character Hamlet said to his friend, “There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy”. Perhaps biota in the oceans gobbling up 12C as part of the oceanic/atmospheric exchange explain the imbalance. Perhaps slowly evolving diffusion impedes disposal processes from removing the 13C in direct proportion to 12C. There are any number of possibilities.
The most you can say is the the 13C/12C ratio is consistent with your hypothesis. But, the rumbling of the volcano is consistent with the gods being angry. Consistency is not proof.
‘In “my” theory there are only two assumptions…’
Plus, many assumptions relating to how the signals are shaped and blended together to remove the trend from temperature and combine temperature related and anthropogenic input together in just the right measures.
Every assumption you make in a chain which depends on the previous assumption dramatically reduces the probability that you are correct, as probabilities combine geometrically. Thus, the odds of getting a coin toss come up heads is 50%. The odds of two in a row is 25%. Three, 12.5%, and so on. This is the basic foundation of Occam’s razor, the recognition that increasingly complicated models become geometrically less likely to be true.
Furthermore, the more complicated the model, the more parameters you have to spuriously reduce the variance with observations. As Von Neumann said, “With four parameters I can fit an elephant, and with five I can make him wiggle his trunk.”
I only require one parameter: the scale factor, and both the slope of the trend and the variation agree. That is not luck. The odds are very much in my favor.
Bart,
You need two control knobs to match the slopes and/or the amplitudes: a factor and an offset. If you match the slopes with a factor and the slopes are not too far different, the amplitudes more or less match, but if the slopes are quite different, like a non-increasing temperature or a negative temperature, the amplitudes go astray, which is the case in over half the full period. Even in your example, the calculated amplitude is only halve the observed one…
I need two control knobs, one to match the amplitude of the variability and one to match the slope. The latter is simply empirically deductible from the net sink rate caused by the extra CO2 pressure in the atmosphere above steady state for the ocean temperature. The rest is straight forward calculation for any first order linear process. Not one extra assumption needed. And no need for temperature as main cause of the CO2 increase, which it can’t be…
Yours is simply curve fitting, not based on any physical process (to the contrary, it violates Henry’s law for the solubility of CO2 in seawater). Mine is directly based on established physics…
No, to match the slope, you only need one. The offset does not affect the slope.
“… which is the case in over half the full period.”
No, it is not. A least-squares-fit trend line through short, noisy segments of data is not “truth”.
“Yours is simply curve fitting…”
Yours is simply more complex curve fitting, and the added complexity makes it more likely wrong.
“… it violates Henry’s law for the solubility of CO2 in seawater”.
No. This is not a static process. You are misapplying Henry’s law.
Bart,
You need a factor ánd offset to match the slopes or the effect of temperature is simply zero on CO2 levels over the full period.
The residual of human emissions for the pCO2 difference with the seawater temperature is midst the variability, assuming a linear process of removal of any CO2 above dynamic equilibrium, with an empirical e-fold rate of over 50 years:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/dco2_em6.jpg
The transient CO2 response to temperature variability is exactly the same as from the temperature, but has near zero trend:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/had_co2_nat_deriv.jpg
As that are two quasi-independent processes, the combination of both gives at least an equal good fit to the observations of trend and variability as your “temperature fits all”…
Figured I’d take a look back to see if anything new or useful had been said. Nothing has.
“You need a factor ánd offset to match the slopes or the effect of temperature is simply zero on CO2 levels over the full period.”
A “slope” is the amount of rise divided by the time over which the rise occurred. I know we have both been sloppy with the terminology at times, but I am using the precise meaning here. The slope is completely independent of offset.
“As that are two quasi-independent processes, the combination of both gives at least an equal good fit to the observations of trend and variability as your “temperature fits all”…”
Not really, and with a great many more assumptions and complexity. Sometime, if I ever get around to it, I will show you why it is not as good. But, it is rather complex, and I feel little motivation to wade into complex discussions with you, Ferdinand, as you do not have the background, and whatever effort I put into it will likely be wasted.
Bart,
I agree, you need only a factor to match the slopes, but you need a factor and offset to match the full effect of temperature on the CO2 rate of change. Still the amplitudes are only halve the real ones, simply because you need a different factor for amplitude and slope, as these are proven from different processes… That will match the amplitudes, regardless of positive or negative slopes in parts of the trend…
Not really, and with a great many more assumptions and complexity.
Again, there are only two reasonable assumptions: most of the CO2 rate of change is a matter of transient response (of mainly vegetation) to fast (1-3 years) temperature changes and most increase in the atmosphere is the response of nature (mostly the oceans) to the CO2 increase in the atmosphere above steady state per Henry’s law. That is all. Simple, straightforward calculations. No further assumptions were made or necessary.
Sometime, if I ever get around to it, I will show you why it is not as good. But, it is rather complex, and I feel little motivation to wade into complex discussions with you, Ferdinand, as you do not have the background, and whatever effort I put into it will likely be wasted.
Much of the mathematical background is quite difficult to follow for me, but it is also clear that you lack the physical/chemical background in ocean carbon chemistry, or you wouldn’t come out with (and defend) something like CO2 piling up in the ocean surface, which is physically and chemically completely impossible with warming oceans, except if coming out of the atmosphere, not reverse…
Michael Hammer,
The integration of the noise in the CO2 signal gives zero to a slightly negative trend after 1-3 years: the noise is caused by the reaction of tropical vegetation on temperature changes (Pinatubo, El Niño), but vegetation as a whole is a small, increasing net sink for CO2…
Ferdi, your one size fits all figure outgassing does not match the data:
http://climategrog.files.wordpress.com/2013/11/co2_cruise_map1.png
https://climategrog.wordpress.com/co2_cruise_map-2/
Mike,
That is from only one cruise. There are many others and there are time series from fixed stations (Bermuda, Hawaii,…). The compilation by Feely e.a. shows the yearly average release and uptake zones for the reference year 1995:
http://www.pmel.noaa.gov/pubs/outstand/feel2331/maps.shtml and next section.
This map yields an annual oceanic uptake flux for CO2 of 2.2 ± 0.4PgC/year for the yearly weighted average CO2 difference between atmosphere and ocean surface including the sinks and sources to/from the deep oceans…
“It matches the slopes because you used an arbitrary factor and offset, which always can match any two linear slopes…”
No, Ferdinand, you cannot match the trend with an arbitrary scale and offset. That would cause a mismatch with the variation. It fits so well because both the trend and variation are from the same process, and the scale factor and offset that match both of them are the same.
“Because you don’t like the result, it must be “manipulated”?”
No, because it required so many arbitrary assumptions, and eliminated the trend that was already in agreement between the rate of change of CO2 and the temperature.
“Have you any data which shows that “my” theory, based on real world processes, doesn’t fit any observation?”
Yes. Your theory fails because it arbitrarily eliminates the trend in dCO2/dt and T which already match one another very well.
Wrong thread. Ignore.
Through an innovative procedure the author reached global warming component as 0.2 oC per Century, though it is illogical logic like cow is white and wall is white so cow is wall or wall is cow. It surprised me, as through the normal procedure I reached 0.1 to 0.15 oC per century — 1880 to 2010 — as global warming component. Urban heat island effect and its associated heat advection in the wind ward direction [wind rose and inversion layer height define this] and rural cold island effect and its associated cold advection in the windward direction. In the cold island effect, temperature increase is not directly proportional to humidity increase. It depends upon the cloud formation and giving rains – thundershowers or otherwise–, etc.
See Solar Energy, 38:97-104 [1987] “The estimation of global solar radiation and evaporation through precipitation – A note” – cube root of precipitation related to global solar radiation and evaporation; Solar Energy, 15:281-285 [1974] “An empirical method for estimating sunshine from total cloud amount”. Also, presented articles relating the square root of precipitable water and net radiation with the wet bulb temperature and that in turn through relative humidity, square root of pressure and temperature. Square root of saturation vapour pressure [SVP] at mean temperature is related to global solar radiation. Global solar radiation and net radiation follow 10.5±0.5 year cycle and its multiples. So, it is not that simple to say humidity vs temperature vs global warming. In addition to these wind [speed & direction] plays important role – that is general circulation patterns of the region.
All these play the role in reflection, absorption and re-radiation and thus changing the radiation balance and thus change in temperature.
Finally, in nature with so much variations, such curves [one to one with low deviations] are rarely possible.
Dr. S. Jeevananda Reddy
Dr. S.
One small mistake. My estimate of 0.2 Deg. C was an approximate upper bound on the amount of heating that might be forestalled by eliminating the burning of fossil fuels. However, it seems that you only used my estimation as an entree for your cow ‘logic’ and thick prose.
CO2 and temperature
Correlation isn’t causation
Particularly if you are only using data for 56 of the last 4.2 billion years
Physics alone tells you this.
If you increase C02 and hold everything else equal, then the temperature will increase.
That’s been known for over 100 years.
But nobody knows how much
only if you can hold everything else equal.
The physics alone also shows like below.
The main inconvenience is like adding a drop of water each day to a swimming pool, how long until anybody evens notice there has been a change?
If you don’t keep external influences away then evaporation and rainfall are much large influences.
The significant cooling of the oceans between 40-60N and 40-60S show CO2 is having no influence on them. In fact these regions have cooled so much they might be even not far off those during the LIA. Once the warmer Tropical oceans due to more solar energy at the surface have cooled, the global temperatures will plunge the way the planet is behaving recently.
I haven’t read all the comments yet so maybe someone has said this.
Perhaps the correlation is backwards. Maybe the “increase in CO2/temperature” is not caused by increased human population. Perhaps the human population has increased because of the unusually temperate climate that has been fairly optimum for human growth and development with brain development that allowed humans to take advantage of fire, clothing, and fossil fuels. Perhaps it was climate change that changed humans and not the other way around.
Now it can become a circular argument, but what started it all? Conditions had to be right before humans had any influence. The Beagle has sailed.
How does one best post an image here?
[Reply: Just cut and paste the image address. Don’t use tags, like ‘[img]’, just use the address. Any link that ends with ‘.jpg’, or ‘.gif’, or ‘.png’ will show the image. Others might not, but readers can still click on the link to see the image. ~mod]
I have a photo of tree rings that demonstrates variation of up to 6 x in growth rate/year in a particular environment. Convert one tree into forests and what is the net effect on carbon uptake? Noise?
Thanks
http://s569.photobucket.com/user/michael_carter13/media/DSC04280_zpsx8fgfm9p.jpg.html?filters%5Buser%5D=143444870&filters%5Brecent%5D=1&sort=1&o=0
And if the link is .png but the page is html, chaos will ensue.
http://files.ratamill-co-nz.webnode.com/200000436-e4afbe5a79/DSC04280.JPG
Gottcha! (I hope)
So, this represents the last 30 odd years in a sub-tropical location. Rainfall throughout the last 40 years has varied between 55 and 75 inches/a. All other physical variables aside from temperature has been constant. I suspect the variation is primarily due to rainfall during the summer months. The message is: don’t overlook the rapid response of biota as feedback mechanisms.
Quote 1. “Remember that correlation does not mean causation!”
Quote 2. “One needs to ask why OCO-2 isn’t confirming the presumed Northern Hemisphere anthropogenic CO2 when it is blamed for the historic temperature increases, and why there is a larger increase in high-latitude temperatures where CO2 has the lowest concentrations!”
Quote 3. “In summary, approximately 81% of the warming in the last century may have resulted from all anthropogenic influences, as suggested by figure 4.”
The author says not to confuse correlation with causation, then comments that the data evidence actually doesn’t fit the AGW model, then goes on to draw the final conclusion based on the assumption that correlation does mean causation and the AGW is right after all!!!
The data sets reported start with population levels of 2.5 billion – this is 1950. Sixty five years is not long enough to draw any causation conclusions at all given the massive hockey stick shape of the human population increase curve and the total lack of fit of global temperature and human population prior to 1950.
The author was correct in his first statement.
Bernard,
The operative word in quote 3 is “may.”
Thanks Clyde. I will give you that. Kudos for all your replies to the responses to your post by the way.
Bernard,
Thank you. I’ve taken a lot of abuse, some of it petty and trivial. Some of it even ‘friendly fire!’ There seems to be two camps that have dug in their heels and don’t want to hear anything that isn’t along the party lines. My goal was to point out some problems with the ‘accepted’ views and try to get people to think and continue questioning so that a coherent, supportable hypothesis can be put forward. Secondarily, I was trying to hang supporters of COP-21 by their own petard by pointing out the consequences of their assumptions. Perhaps that was too subtle for some.
Clyde, you don’t do anything good when your arguments are based on shoddy analysis. If you want to do good work, review how Willis goes about looking at things!
v/r,
David Riser
Clyde, since you felt it Important up thread to correct a poster’s grammar, I feel I should point out that you can not be ‘hung on a petard’! The correct verb is ‘hoist’.
Phil,
Thank you. I actually learned something today. I’ve never actually read Hamlet in its entirety.
Note that I don’t bestow the honor on everyone. David was a special, deserving case.
Well Clyde if all you can do is take me to task for grammar, then I guess you don’t have anything to say about the quality of your work!
v/r,
David Riser
David,
There is an old saying that “There are none so blind as those who WILL not see.” So, let me make it very clear: You arrogantly denigrated my statistical skills by saying that I needed remedial instruction. Yet, you demonstrated that the crucial skills necessary to critique others was wanting, at the basic level, by yourself. You also demonstrated that you could not learn from constructive criticism.
As to the validity of your complaints, I have discussed it with my neighbor across the street who is a senior mathematics professor at the Air Force Institute of Technology. He does not share nor support your mathematical criticisms.
Quality addressed. End of discussion.
Well Clyde your tone has not been particularly kindly. I would say that at times I get a bit cranky when statistics is obviously being abused. Followed with a correlation doesn’t equal causation but look at my shiny numbers. Reality is that CO2 is such a small contribution to global temperatures that your numbers are essentially impossible from a physics perspective. CO2 was measured in Feldmen et al (2015) “Observational determination of surface radiative forcing by CO2 from 2000 to 2010”. While their conclusions about multiplying the CO2 effect by 3x due to WV feedbacks have not occurred, they did measure an increase over 10 years of .2 +-.06 W/msquared. which is essentially nothing when you consider the typical clear sky measurement of SDWLR can range from close to zero to 500 w/m2 depending on temperature, water vapor, location etc. a good reference for this is found here http://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20080000844.pdf.
So the idea that an increase over 10 years of .2wm2 could account for over half the warming is ludicrous on the face of it.
v/r,
David Riser
I would like to propose a more suitable dependent variable: time.
1st, CO2 is going up monotonically as we produce it faster (that is, the human 5% of contributions) than it is removed (through things like limestone creation).
2nd, temperature is still going up out of the last glacial period, absolutely naturally that is. It is possible to see it in historical temperature records even before we started burning coal.
The apparent correlation between temperature and CO2 is simply a common dependency on time.
Hivemind,
That is, of course, the potential “coincidence” that I acknowledged that one has to consider. But time isn’t ‘causing’ the increase in either variable. There is something else, such as the Milankovitch parameters, which are changing with time, that are the drivers behind the change in temperature. That being the case, then temperature would be the independent variable and CO2 the dependent variable. However, there is no clear evidence of what the actual driver(s) is/are. The current IPCC paradigm is that CO2 is the independent variable driving temperatures. The point of my scatter plots was to point out that what appears to be a good correlation between temperature and CO2 in recent times breaks down as one goes back in time. That suggests that either the historical data are unreliable or that the relationship is not as robust as it appears in recent times.
A bit late here but I have to say it. Limestone is quite possibly being destroyed at a rate that exceeds its rate of formation, Ever drink the water in a limestone district? It tastes good, because of the high lime content (of course that may be my personal taste because I grew up in a limestone district). Limestone is dissolving in acidic rain water, all around us, all the time. And it produces a heck of a lot of CO2.
Any attempt to quantify natural sources and sinks, is little better than a guess. The only CO2 parameter that can be estimated within any reasonable error limits is that from fossil fuels, because they are bought, sold and quantified.