Guest Post by Willis Eschenbach
A recent post by Anthony Watts highlighted a curious fact. This is that records of some two and a half million oceanic pH samples existed, but weren’t used in testimony before Congress about ocean pH. The post was accompanied by a graph which purported to show a historical variation in ocean pH.
I was unimpressed by the graph in that post, which seemed simplistic and, well, in a word, wrong. But on the other hand, I certainly found it bizarre and most interesting that someone would throw out that huge amount of scientific data. That was the reason I forwarded it to Anthony, in the hope of unraveling the actual truth of the matter.
So … as is my wont, I’ve now taken a look at the data myself, albeit at the moment a very preliminary look. The data was conveniently provided by a WUWT commenter in .csv format here, my compliments to him for the collation. He also has a good explanation of the process, along with R code. Note that there has been no quality control on the data. About 2% of the surface pH values are well outside the range of oceanic pH, and I removed them before looking further at the data.
Now, the first question I asked was, where were the samples taken? The problem with the graph in the recent post linked to above is that it lumps together samples taken in various parts of the planet. And unless the sampling is uniform in time and space, this is a Very Bad Idea™.
So I made a map that shows where each surface sample was taken. For simplicity, and because this was my first cut, I restricted myself to those samples with a depth of 0 (right at the surface), which are a bit less than a tenth of the total samples. Here are two different views of the same location data.
Figures 1a and 1b. Two views of the location of the surface samples of the global pH dataset, centered on the Pacific and the Atlantic. In some regions you can see the tracks of the oceanographic expedition vessels quite clearly.
Now, I must confess that this was a surprise to me. I hadn’t expected the concentration of samples around Japan, it appears the Japanese oceanographers mush have been quite busy. And I also hadn’t expected the high sample density in the Baltic Sea and the other enclosed seas (the Black Sea between Turkey and Russia, and the Caspian Sea to its right).
Finally, here are the average pH values by gridcell, for the entire period of record
Figure 2. Average values of pH by gridcell in the record.
Now, you can see from these maps that we cannot simply put all of that data into a single box and extract a timeline from it.
So … was there “pHraud” in not utilizing this data? I say no, there was no fraud. I say this in part because it’s so difficult to infer intent. Because I have been falsely accused of having bad intent a number of times, I’m sensitive on the subject. I dislike accusations without evidence, and I see no evidence of fraud in this case.
However, it is a huge scientific resource, two million plus pH samples taken by oceanographers over decades, and not using it without some solid scientific reason for ignoring it just doesn’t work for me. What I suspect has happened is that the mass and complexity of the data was too overwhelming, and so the investigators simply put it into the “Too Hard” pile. But that’s just speculation, the real reason may be entirely different. Regardless of the reason, I do think that the authors should have explained their omission.
In any case, that’s the story so far. It certainly appears to me that there is plenty of data there for meaningful time series extractions in some areas. There are, for example, about 400 1°x1° gridcells that have more than a hundred observations per gridcell, and groups of nearby gridcell cells combined have much more data. The North Atlantic and the oceanic area off of Japan seem like they would have more than adequate data for time series extraction.
I may or may not do any followup on this dataset, but I invite readers to use the data for their own analyses.
Regards to all,
w.
ADDENDUM: As usual, I request that if you disagree with someone, please have the courtesy to QUOTE THEIR EXACT WORDS THAT YOU DISAGREE WITH, so that we can all understand the exact nature of your objections.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
What about dates? The assertion is that the pH is decreasing, right? So there has to be a trend. Second, a flat average is meaningless. The data has to be kriged, and from the look of the graph, there are (as usual) huge regions that are undersampled and smaller ones that are heavily oversampled.
rgb
Yes, dates are essential. The authors didn’t say the earlier data didn’t exist. They declined to cite an annual average. That would mean they believe there wasn’t enough good data in that year, with adequate spatial coverage, to compute an average. That is what has to be checked.
GISS gives a global temperature average starting 1880. There is plenty of data from some places before 1880, but they don’t think the coverage is good enough for a global average. No one seems to think that is improper.
“Note that there has been no quality control on the data. About 2% of the surface pH values are well outside the range of oceanic pH, and I removed them before looking further at the data.”
Sounds like an important caution. Sabine et al may know stuff about data quality that experts here don’t.
Nick–
This graph seems to show that for the period from 1960-1980 when there were a large number of measurements there was small annual variation in pH, but both earlier and later when the number of measurements dropped off there was more variation.
https://dl.dropboxusercontent.com/u/75831381/pH%20vs%20time%20and%20number%20of%20obs.pdf
Nick, “No one seems to think that is improper.”
I think that is improper. I’m not “no one”.
https://chiefio.wordpress.com/2011/07/01/intrinsic-extrinsic-intensive-extensive/
ANY “Global Average Temperature” is an absurd and non-physical number. Every single one.
So all you can do is FIRST and FOR ANY ONE THERMOMETER compute an anomaly trend; then average those trends and talk only about anomaly trends. (I hear 1000 keyboards preparing to say that is what is done… it isn’t.) Now realize, that means you must NOT average the Min-Max per day, and NOT average the days per month. Then NOT carry those monthly averages AS “temperatures” through the rest of your homogenizing (as GIStemp does) and only calculate a “grid cell anomaly” at the end (as GIStemp does).
Given all that machination, and all that fundamentally dumb (i.e. irrational) averaging of an intrinsic property (that can not be averaged and retain meaning as a temperature) it is highly irrational to ‘cut off’ the series at any point, or to use it with any variation in instruments in place (and expect more “averaging” to fix the induced discontinuities). It is fundamentally improper to use ANY variable set of thermometers and average them.
For a bit more depth:
http://www.uoguelph.ca/~rmckitri/research/globaltemp/globaltemp.html
http://www.uoguelph.ca/~rmckitri/research/globaltemp/GlobTemp.JNET.pdf
Ross is not “no one” either…
Per pH variation
It is also essential that Time Of Day and level of sunshine is known. Daily pH cycling in surface ocean levels can be fairly large based on relative photosynthesis / respiration of the stuff living in it… So if you do not know the TOD profile and the insolation levels, you might be only indirectly measuring sunshine and fertilizer…
rgbatduke December 30, 2014 at 2:30 pm Edit
Nick Stokes December 30, 2014 at 4:13 pm Edit
Dang, guys, take a deep breath. You might want to re-read what I said at the onset:
So yes, the dates are important. And yes, a more detailed quality control is important … but it’s a PRELIMINARY LOOK.
Next, a flat average is not “meaningless”, as it allows us to see what the average pH is (according to their measurements) in various areas of the ocean.
Finally, if you want a better analysis, re-read what I said above:
I’m just moving the project down the road. I made no claim as to its completeness, just increasing our understanding.
w.
Well, I’ve downloaded the data. I allowed down to 1 m depth (not just 0) and got 202354 points (min pH 0). I did a table over years. No QC.
I don’t think this was Sabine’s data. There is very little after about 1990; it’s mostly between then and 1960. Best year was 1978 with 7106 total.
Now even if that was uniformly distributed, it would be far too little. People here complain about the reduced size of GHCN. That has about 1800 stations with about 12 readings per year. And they are reasonably well distributed. Here we have at least double the area, and at best about 1/3 the points, with obviously eccentric distribution. And then with QC…
[Reset to ASCI “pre” format. .mod]
Thanks, mod – I’ll try to remember that.
Nick Stokes’ table of pH measurements by year shows a sharp drop in the number of surface measurements from about 5800 in 1988 to 1500 just 4 years later and down further to the ridiculously small number of about 150 by the year 2007. Strangely enough, this is the exact period (1989-2007) chosen by Feely to illustrate the drop in pH:
http://wattsupwiththat.com/2014/12/23/touchy-feely-science-one-chart-suggests-theres-a-phraud-in-omitting-ocean-acidification-data-in-congressional-testimony/
The graph of pH vs year by Mike Wallace also shows a sharp drop, more like a step function, in pH values between about 1989 and 2004, again coinciding with the major reduction in the number of measurements:
https://docs.google.com/document/d/1ijyZwW2e_OJEwJiimgnO-UF9WF_1k_2GWfnhYThq2O0/edit#heading=h.1jqf6pash65r
If one wanted to make the case for ocean acidification, there is really only one place to do it, and that’s from 1989 on.
Lance W,
“Strangely enough, this is the exact period (1989-2007) chosen by Feely to illustrate the drop in pH:”
Yes, but the dataset is the one chosen by Mike Wallace, with all the “omitted data”. It’s unlikely to be what Feely was using for the post-1988 period.
” People here complain about the reduced size of GHCN. That has about 1800 stations with about 12 readings per year. And they are reasonably well distributed. ”
Ahaha! Really, how many of them are in the middle of oceans?
Nick Stokes:
I write because it is rare for me to agree with you so I welcome this opportunity to do it.
You say
Emphasis added by me: RSC
I very strongly agree and I have repeatedly made the same point about this data – including making it on WUWT – whenever Ferdinand Engelbeen has cited this data.
The data is far too sparse in spatial distribution and over time for it to be meaningful as indicator of the putative alteration of average oceanic pH resulting from increase to atmospheric CO2 concentration over the last century.
Richard
Lance, the graph of the other Wallace is solely based on glass electrode pH measurements, which are near completely abandoned in recent years as too unreliable (with a lot of care not better than 0.1 pH unit) to measure a general pH drop of theoretically 0.1unit since 1850 (with about 50% of the CO2 increase since 1984 in the atmosphere) since the start of the industrial revolution. Moreover his pH drop of 0.3 pH unit since 1984 is not the result of a global or local pH drop, it is the result of the random sampling in different places and different seasons.
The curves by Sabine, Feely, Bates and many others are based on either calculated pH, or pH measurements based on far more accurate colorimetric pH measurements or both. They excluded glass electrode pH measurements as too unreliable.
pH can be calculated by (over 80 years) established ocean carbon chemistry from other, better defined measurements like TA (total alkalinity) and DIC (dissolved inorganic carbon).
See: http://www.pnas.org/content/106/30/12235.full.pdf
for a description of the methods and the trends at Hawaii, where both methods were used.
Before 1992 direct pH measurements were by glass electrode:
http://hahana.soest.hawaii.edu/hot/methods/ph.html
but not used in their graphs
For Bermuda:
http://www.biogeosciences.net/9/2509/2012/bg-9-2509-2012.pdf
computation of pH is mentioned in chapter 2.7
More data for several fixed stations can be found at:
http://www.tos.org/oceanography/archive/27-1_bates.pdf
Then the data for regular sea cruises by research ships, here for Japan:
http://www.data.jma.go.jp/kaiyou/english/oa/oceanacidification_en.html
And there is an interesting PH.D. work about the results from cruises in the Southern Oceans which tries to combine pH measurements with calculated pH over time (-0.03 +/- 0.03 pH unit since 1972):
http://scholar.colorado.edu/cgi/viewcontent.cgi?article=1456&context=honr_theses
which contains many interesting points and graphs. Especially the graphs of the distribution in time and latitude of the measured and calculated values (page 46) is interesting.
More to learn about carbonate chemistry in the oceans in a nice overview:
http://www.soest.hawaii.edu/oceanography/courses/OCN623/Spring2012/CO2pH.pdf
Richard,
The compilation by Wallace is unreliable, because he didn’t sort on place and seasons. Regular cruises over the same tracks give a better insight, even if done with less accurate equipment. It is better to recalculate pH from other, more frequently measured variables like TA and DIC, but that is a hell of a job to sort out.
Anyway, you don’t need to monitor all parts of all oceans frequently each year, as ocean chemistry is the same everywhere and the changes caused by (seasonal) temperature (and accompanied bio-life) changes level off after a full seasonal cycle. Ocean pH is directly influenced by atmospheric CO2 levels, which influence at any place of the oceans can be calculated if a few local variables are known. It would be quite strange if none-monitored open ocean places would show an increase of pH with increased CO2 in the atmosphere (far more variable at upwelling places and estuaries).
All fixed stations and repeated cruises show a declining trend since 1984. All in ratio to increasing CO2 levels in the atmosphere.
A compilation of Southern Ocean cruises (31,700 measured + 41,800 calculated pH values) also shows a declining trend since 1972.
(links are currently under moderation, will be shown soon).
Ferdinand Englebeen said:
I find this difficult to believe, Ferdinand. I know the salinity of the ocean varies quite a bit from location to location. And although salinity does not directly affect pH, I can believe the chemistry of the ocean that does affect pH also varies quite a bit by location.
What appears to be fraudulent about the Feely/Sabine study is that they apparently failed to mention that they had used mostly modeled data and that only the data since 1988 were measured. Had they been a bit more forthcoming in their study, they would have mentioned the existence of the full set of measurements and provided their rationale for excluding it. They also would have explained how they got the data they did provide in their time series plot. With the history of the IPCC climate models, there is good reason to have serious doubts about any modeled data.
Ferdinand, thank you for the valuable references. I note that the NOAA website used by Simon Filiatraut to download 2.4 million pH measurements
(http://wattsupwiththat.com/2014/12/23/touchy-feely-science-one-chart-suggests-theres-a-phraud-in-omitting-ocean-acidification-data-in-congressional-testimony/#comment-1823345)
also offers various related measurements, including temperature, salinity, alkalinity, CO2 pressure, oxygen, etc. As you point out, these could be used to calculate an expected pH and therefore could serve as a quality control check on the pH values.
Although the carbon chemistry may give an indication of the effect on ocean pH of increasing CO2 in the atmosphere, much may also depend on the biological and geological feedback, so I am not convinced that we can estimate the future effect very well.
Lance, the data from NOAA were further gridded Berényi Péter in 5×5 deg boxes and found near zero trend over the past 30 years for all gridded boxes. The error margin still is large +/- 0.03 unit per decade for the glass electrode measurements, where the modern measured/calculated methods show a trend of around -0.015 unit per decade, largely within the error margins of the glass electrode measurements.
What we already can conclude is that the glass electrode pH measurements can’t show the trend over the past 30 years and that the compilation of the other Wallace doesn’t show a real pH trend, but the result of random sampling in place and time.
Indeed it should be possible to use the other historical measurements to calculate the pH of that moment. The last reference I sent was for a PhD work which has done that for all known data for the Southern Ocean. Seems a lot of work, but maybe programmable now that all data are in computer readable form…
Apart from the question of a trend, there are a few other aspects of the data set that seem of interest.
1. There is an odd rather symmetric relation of the measured pH to latitude, reaching a maximum of 8.1 at about 30 degrees N and S, dropping back to 8.0 at the equator, and also dropping below 8 toward the poles.
2. The later years (1988 on) not only show a drop in pH, they also show a sharp increase in the median depth, fromd about 200 m to about 1000 m. And pH showed a significant Spearman correlation of -0.15 with depth, and a highly signficant though small Pearson regression coefficient of -0.01 per hundred meters. This might have played a small part in the Feely Sabine calculation dealing with these same years.
3. Ferdy E., what direction do you expect the seasonal variation to take? In both hemispheres, it appeared that the lowest values were in summer and the highest in spring.
The data is far too sparse in spatial distribution
=============
sort of like using Mona Loa for global CO2?
Ferdinand…their fig 4 pH from your Hawaii link….just looks like a sine wave…no steady decrease in pH at all
honestly, it looks like if they had started in 1979 and stopped in 2007, they would have shown a huge increase in pH
Looks like a temp graph where you cherry pick your time to show anything…and frankly, seeing it that way makes it clear that any wiggles in pH, either up or down, is just nothing more than that…..a wiggle
http://www.pnas.org/content/106/30/12235.full.pdf
Doesn’t krigging assume a smooth spread? Is there any reason to believe the ocean pH is properly represented by a smoothed even spread? a lack of “hot” or “cold” spots? spots of upwelling acidic or basic water from underwater features or events? should these be kirgged away? or krigged over?
Does the ocean even have “an average pH” that could be going up or down? or is this just another invented concept created for propaganda purposes — world in danger or world resilient? is it just another weapon in the Climate Wars?
Kip Hansen
You ask
I provide my answers to each of your three questions in turn.
Q1
Does the ocean even have “an average pH” that could be going up or down?
A1
Possibly, but probably not
(see comment by Lance Wallace at December 31, 2014 at 3:53 pm).
Q2
is this (i.e. “an average pH”) just another invented concept created for propaganda purposes — world in danger or world resilient?
A2
Yes.
Q3
is it (i.e. “an average pH”) just another weapon in the Climate Wars?
A3
Yes.
Richard
The gridded result:
Berényi Péter December 31, 2014 at 8:58 am
A preliminary look at the data shows beyond doubt, that ocean pH is decreasing indeed, at a rate of -0.002±0.038/decade. In other words, it is absolutely stable.
delivers virtually the exact same answer as simply treating the data as a random sample, without any gridding, averaging, anomalies, adjustments, etc.
ferdberple December 31, 2014 at 8:29 am
http://oi60.tinypic.com/9s7xvo.jpg
I’m sure I read a while back that if you record the PH in the morning and again in the evening, in the same place, the samples will show different data.
That is true. There are many physical occurrences that impact pH … like surface agitation, for example.
Seawater pH changes with temperature, time of day, season, depth, light level, whether the sample is filtered or not… Diurnal variation in “A Shallow Texas Bay” as reported in RA Horne’s Marine Chemistry, 1969 was ~0.7.
I agree, Norm. Actual pH will vary considerably due to temperature, salinity, calcium carbonate content, etc., which makes some previous comments about the chemistry of the ocean being the “same everywhere” suspect, IMHO.
One point I have not seen yet is the impact of the temperature compensation of electrode pH meters. Many meters have both electrode temperature and actual temperature compensations that are often set to correct the reading to 25C. I have no idea of what type of data we are getting. However, I know I would like to see the actual pHs, because that is what the sea life is encountering.
Last point about “average pH.” You might think the average of pHs 7.0 and 8.0 is 7.5. But due to the logarithmic nature of pH, the average is actually 7.26. So unless your averaging program first converts the pHs to actual H+ ion concentrations, then averages them and takes the log of the results, then you will not have accurate averaging.
Can the data be resolved into decadal trends? In addition to “where”, “when” the sampling was conducted is crucial. The lower pH around the Bering Sea and northeastern Pacific could easily be dominated by upwelling events that are modulated by the PDO. Average pH could be skewed by the time of sampling.
Thanks, Willis.
I agree that throwing out so much data needs explaining,
Maybe it just went overboard?
Happy new year!
Could you please provide the cut off limits for the data considered outside normal sea pH?
Willis
The methods for taking co2 samples since 1820 And up to the time that keeling started Sampling at mauna loa have been roundly criticised. Is the method for taking these ph samples considered robust and consistent bearing in mind the number, the time scale and the variety of people taking them?
Tonyb
good grief! The gremlins ate much of my comment above.
To paraphrase, bearing in mind the complaints about historic co2 readings that were regularly taken from 1820 up till Keeling changed the methodology in the 1950’s, can we be certain that these historic ph readings are robust?
tonyb
Hi Tony, ferdberple does show a plot of all sampled pH data over the years near the end of the discussion. Has a similar look as the cloud of CO2 data from the many samples taken in the past… pH data ranging from zero to 14???
I agree with Sabine: most of the data are not reliable but may give some idea if sorted for place and season…
Hi Ferdinand
Having just seen your post I have given my reply to fredberple here
http://wattsupwiththat.com/2014/12/30/ph-sampling-density/#comment-1825726
Tonyb
samples around Japan….whaling is scientific for them
8.25….higher than the buffer…..plankton??
There’s way too many 8.25’s….what’s up with that?
Willis, can you divide this from the time of formula measurements…and actual measurements?
8.25 is very close to the ideal buffering point, as illustrated here: http://water.me.vccs.edu/courses/env211/changes/carbonategraph.gif
That is of course why buffers are used to maintain pH close to a fixed point.
except that looking at the gridcell map Willis posted…..it’s not possible where it’s shown
pH over 8.25 is not unusual in areas with high biologic activity where algae “eat” the CO2. Over coral reefs pH can touch 9.0 in late afternoon.
Mmm… I don’t think you will get very far asserting that Japan is a low-tech nation. I would advise eschewing that approach.
no one said that Bart???
Then, I misunderstood. The meaning of the remark wasn’t clear to me. My apologies.
oh, I think I see what it was….
The Japanese use “scientific research” as an excuse for whaling, tuna, fisheries…
…so their commercial fishing fleet takes measurements every time they go out
That’s why the sample density is so high there..
Better??
Yes. Now it makes perfect sense. Good observation.
Willis, there is something fishy in that ocean less-alkaninization.
The average ocean water contains 10,4 mmol/l ca2+ and 2.34 mmol HCO3-
http://www.seafriends.org.nz/oceano/seawater.htm
Obviously the solubility of caciumcarbonate at pH8 is 1mmol at 25C (fig 2):
http://www.ldeo.columbia.edu/~sanpisa/OceanSed%20project/factorscontrolcarbonate.htm
so it appears that the combined Ca2+ and HCO3 – are oversaturated but that is temperature/pressure and pH dependent. It occurs to me that adding additional CO2 would result in the precipitation of CaCO3 as it exceeds it’s solubility, rather than it would lower the pH.
“It occurs to me that adding additional CO2 would result in the precipitation of CaCO3 as it exceeds it’s solubility, rather than it would lower the pH.”
You need to work with the solubility product here. Solubility .001M doesn’t help, because it assumes no excess ions present. The solubility product is that of Ca++ and CO3–. The reason added CO2 dissolves CaCO3 is that, which it adds carbon, it converts CO3– to HCO3-, thus reducing the product.
Carbon dioxide in solution forms a buffer solution with the form taken the carbon dioxide strongly dependent on pH:
CO2 + H2O H2CO3 (carbonic acid) upon becoming absorbed by water
At low pH it forms bicarbonate ion
H2CO3 H+ + HCO3- (balance point pH 6.37)
At intermediate pH bicarbonate forms carbonate
HCO3- H+ + CO3(-2) (balance point pH 10.36)
At higher pH it is predominantly present as carbonate.
At the prevailing pH of most sea water, about 8.3 to 8.4, the CO2 present in sea water is almost entirely present as bicarbonate, with barely 1% of it present as carbonic acid and another 1% present as carbonate ion.
This is good because the bicarbonate form is the only form biologically available for foraminifera, coral, mulloscs and other sea life that forms calcium carbonate structures.
CaCO3 ‘precipitated’ by sea water (eventually forming limestone, chalk, marble, etc.) was formed by living cells.
CA(HCO3)2 = CACO3 + CO2 + H2O
I think you are right, the enormous amount of data accumulated over a long period will take a lot of time to analyse and quite possibly will not come to the conclusion that the analysts would prefer.
On the other hand, there would be a PhD for someone with the integrity to do an honest breakdown and dissection of the data – assuming that they could find an academic establishment to review the result dispassionately.
I know of no academic establishment that is not a hotbed of activist bias. Individual researchers, perhaps, but not the average student.
There was a PhD work done on part of the data for the Southern Ocean (2012), partly on measured, partly on calculated data:
http://scholar.colorado.edu/cgi/viewcontent.cgi?article=1456&context=honr_theses
I suppose that he used all available data for the Southern Ocean, as that was very sparsely sampled in the past…
The result, with large margins of error: -0.3 +/- 0.3 pH unit over the period 1972-2011
This was a ‘senior thesis’ paper in receiving an undergraduate BS degree, and not a PhD. First couple of lines is all one needs to read to understand the direction of paper. Granted, most ‘senior thesis’ never come anywhere close to this attempt, but one also only needs to look at the principal ‘contributor’ as well (likely thesis adviser).
The Japanese have cultivated seaweeds for centuries and modern operations would be very interested in water conditions, including pH.
http://www.seaweed.ie/aquaculture/noricultivation.php
An interesting side note to the nori cultivation story is the honoring of Dr. Kathleen Drew-Baker whose research on the life cycle of nori (Porphyra umbilicalis) enabled better production and quality. The Japanese erected a statue of her and publicly remember her work over sixty years later. https://en.wikipedia.org/wiki/Kathleen_Mary_Drew-Baker
“Now, you can see from these maps that we cannot simply put all of that data into a single box and extract a timeline from it.”
The same procedure used for the “global temperature” trend?
Go Willis go. This is (another) very important issue that needs an objective mind.
Another issue is how long were sample maintained before pH was determined. Samples were often taken during cruises and then analyzed weeks to months later. There was no systematic timeline for nalaysis. Bottles locked away in the dark prevent photosynthesis that could increase pH, but cellular respiration from bacteria would continue and could create a drop in pH.
One should see wider seasonal and decadal variations with seas like the Black Sea that are more ‘landlocked’ than the open oceans. Most of the Black Sea’s bottom waters are devoid of oxygen. I would think that similar such situations be graphed separately, rather than dumping all data into one graph. Also, rates of inflow and outflow would have to be accounted for.
In a local lake in the Gatineau Park, Canada, there is a lake which is a left-over from the last ice-age. It’s called Pink Lake, and the Oxygenated water only goes to 60 feet, below the oxy-xline, ithe water is oxygen free. Scuba diving i9s forbidden there.
http://www.ncc-ccn.gc.ca/places-to-visit/gatineau-park/pink-lake
New dams that have a lot of vegetation in the flooded area also have a poor distribution of oxygen. Below the oxy line there is a high concentration of H2S. If deep water is released from such a dam into the river below, it kills everything in it. Thus it is common to release some of the deep water and mostly top water while monitoring the mix downstream. It is easy to get it wrong.
Good dam management has to address this issue constantly. I suspect there must be a lot of lakes like Pink Lake in other parts of Canada. Being deep helps create such an unwanted boundary.
What I suspect has happened is that the mass and complexity of the data was too overwhelming, and so the investigators simply put it into the “Too Hard” pile.
I aplolgise for not accepting this idea. You have managed to make something of it in a short timescale and I have no doubt that you or someone like you would be able to do much more if time was available. Time was available to the original ocean acidificationists and they chose not to do it. That tells me a lot.
+1 Andrew … if you have the real data then modelling fake data does not cut the mustard. Time is not of the essence.
Andrew Hamilton December 30, 2014 at 3:44 pm Edit
Mmmm … you may be correct. Although I’d add that I work a lot with big datasets, and I’ve developed my own tools to do the mapping and analysis in a short time. So the idea of crunching two and a half million individual records doesn’t faze me, but others might not be so sanguine about the project.
As I said, we don’t know at this point why the data weren’t used. And until we have a clear answer to that question, allegations of fraud are out of line.
w.
Willis says ” We don’t know at this point why the data weren’t used”
I thought we did. Sabine and Englebeen have said words to the effect that the data did not provide the “signal” that was sought, the signal being the decrease in oceanic pH. They wanted that “signal” and they had to fabricate the data to provide it. This was done by hind casting a trend of declining pH via modeling.
You would think that any reasonably competent researcher setting out to look at global oceanic pH would be prepared to deal with large data sets over large areas. If not, how could they expect to produce valid results?
mpainter, thanks for the alleged motives…
There are two problems with the data:
– the accuracy of the glass electrode pH measurements: 0.1 pH unit, while the theoretical trend from 1850 to 1984 was about 0.05 pH unit.
– the lack of data for the same parts of the oceans in the same seasons over a longer period.
The theoretical trend is confirmed since 1984 at seven fixed station at different places over the oceans, based on more accurate direct (colorimetric) pH measurements and calculated pH (which overlap each other).
The theoretical trend is confirmed since 1972 for the Southern Ocean based on direct (glass electrode) and calculated measurements (31,700 measured + 41,800 calculated pH values), be it with large margins of error: -0.03 +/- 0.03 pH units:
http://scholar.colorado.edu/cgi/viewcontent.cgi?article=1456&context=honr_theses
Here the distribution over time of the measurements:
http://www.ferdinand-engelbeen.be/klimaat/klim_img/ph_sh.jpg
no direct measurements in the period before 1990…
If you see any reason that the pH of the open oceans in general wouldn’t show a faint drop with increasing CO2 in the atmosphere in the period before 1984, I am very interested to hear it…
Ferdinand :
The whole problem with climate science is that theory is substituted for facts and facts are invented. If you cannot see the fault of Sabine and Feely’s methods, then perhaps you share those faults. Do approve of the method of presenting model product as observed measurements, giving no hints of how the data was actually derived?
mpainter, I don’t see much problems by presenting modeled data for the period before 1984, because in this case the modeled data are much better than the measured data, which are widely around the model (contrary to climate models), where the (chemical) model is confirmed in the period after 1984.
The only point is that Feely/Sabine should have made it clear that the early part and the “future” part of the graph was from a model, because the observations were not accurate enough.
Hi Willis, I appreciate you are trying to see both sides of the issue, but I find it difficult to imagine a legitimate reason for discarding, not even mentioning the data, and replacing it with a computational model which is based on – what?
I mean, try to imagine this in the context of a high school chemistry class. “Sorry teach, the practical experiment I did produced some difficult to analyse values, so I threw out the data and based my conclusion on a mathematical model of what I think should have happened … yes, I know I didn’t mention the data I discarded, but you see the data was no good… ” – that conversation would have rapidly concluded in an “F” for that assignment.
It might be that different rules apply when you producing reports which influence the expenditure of billions of dollars of taxpayer money, but I can’t help viewing such “lapses” from the perspective of what it would have meant to me, back when I was being taught about science.
Has anyone attempted to calculate just how much carbon dioxide would be required to change the pH of the oceans to any significant extent? It seems to me that given the density of sea water and the volume of that water, compared against the equivalent values for atmospheric CO2 and the fact that CO2 in aqueous solution is a weak acid, never mind the ample buffering in the oceans, the numbers calculated for carbon dioxide requirements would be very large indeed.
Indeed. I did a simple calculation during the BP oil spill in the Gulf of Mexico. It was a tragedy, no question, but the amount of oil spilled there was on the order of one drop per billion (sorry, don’t have the numbers anymore). The oceans are massive, it takes massive amounts to ‘Fundamentaly” change them.
… and why assume that CO2 is the only mechanism that could possibly move oceanic pH towards neutral?
Because.
That’s it. Because.
It’s the CO2.
Because.
Because the increase of total carbon (DIC, CO2 + bi + carbonate) in the ocean surface is measured…
If the pH drop was caused by e.g. undersea or land volcanoes (SO2, HCl, HF,…) the pH would drop and expel CO2 (reducing DIC). If the pH drop is caused by more CO2 in the atmosphere, DIC increases…
That calculation is made and quite easy to do: one can even calculate the pH from the oceans during an ice age (8.367). All you need is the amount of CO2 in the atmosphere: that is in fast equilibrium with the CO2/bi/carbonate system in the ocean surface and the rest is established ocean carbon chemistry of already 80 years ago:
http://biocycle.atmos.colostate.edu/shiny/carbonate/
Sorry, forgot to add: the increase of 110 ppmv (30%) in the atmosphere gives an increase of DIC of about 3% in total carbon (DIC) in the “mixed” layer of the oceans (0-200 m depth), depending of the Revelle/buffer factor. That gives a change of about -0.1 pH unit since 1850. The exchange with the ocean surface layer is very fast (1-3 years), with the deep oceans much slower.
Ferdinand Engelbeen
You say
Well, yes, the calculation is “quite easy to do” and it does give an answer, but the answer is probably wrong (both for the present situation and for “during an ice age”).
“That calculation” assumes that only inorganic chemistry is affecting ocean pH but – in reality – it is observed that biota affect ocean pH, too.
Richard
Ferdinand
I see Willis has also refuted your assertion here and supported it with a real-world illustration here.
Richard
Richard, the data of Willis show a higher increase of DIC than I reported as the real ratio depends of the “Revelle” (buffer) factor and temperature which changes with latitude. Anyway, the 30% change of CO2 in the atmosphere gives the same change as free CO2, but a much smaller change in total CO2 (DIC) and a very small change in pH.
That can be seen for the different stations:
http://www.tos.org/oceanography/archive/27-1_bates.pdf
Fig. 3 and Table 2.
Ferdinand, thanks for that most clear reference, it’s very informative. It upholds my claim that the ocean is a) becoming more neutral and b) at a rate that is pretty meaningless.
w.
I suppose you could use the methods of BEST on this. The methods are published.
I assume the data have been averaged to produce the earlier graph? Given that pH is a negative log of a concentration, how is the averaging carried out?
How meaningful is any such average, especially in a buffered solution such as sea water?
The analysis was done here:
http://www.co2web.info/ESEF3VO2.htm
Acidification is not really possible in a buffered solution like the ocean. Yes you will get local highs and lows but they will not last. The only way you can get a buffered solution to change pH permanently is to use up all the buffering agent. I know since I used to analyze Deionized river water where the organic content buffering the water changed on a daily basis.
Also consider basalt as a buffering agent as well as limestone.
Seems since all the rest of the ‘evidence’ has proven to be junk they are trying to recycle the tired old ‘Ocean Acidification’ again. I though that beast had already been slayed.
+1…thank you Gail for nailing another one
“The only way you can get a buffered solution to change pH permanently is to use up all the buffering agent”
…and you can only do that in the lab
Gail, this is not a conventional buffer system which you’d find in the lab. In a normal buffered solution the pH is close to the pKa value for the buffer, being a diprotic acid carbonic acid has two pKa’s, 6.3 and 10.3, the ocean is nowhere near these! Addition of CO2 changes the equilibrium system of the ocean towards lower pH.
…assuming the buffer is no being constantly replaced
which it is
“The only way you can get a buffered solution to change pH permanently is to use up all the buffering agent.”
That isn’t true at all. Every time you add acid or base, you shift the pH. The rate of change is reduced by the buffer, but not to zero. The formula is Henderson–Hasselbalch:
pH=log(pKa)+log([A-]/[HA])
for acid HA.
The graph of this is called a Bjerrum plot. You can calculate the seawater pH changes here.
As if drawing a parallel between cAGW and homeopathy wasn’t irresistible enough. There is one significant difference though: homeopathy aims to heal people.
Nick, that again assumes all buffering is chemical. Biological activity is not included, and likely changes the result. In addition, if surface winds change over time, the mixing depth changes, and affects the surface average.
Adding acid to a buffer will lower the pH and adding base will raise the pH, but the changes will be less than without buffer. That is basic introductory chemistry. With a well designed buffer (equal amounts of acid and conjugate base and a reasonably high concentration of both) the pH changes can be very small. But the ocean is not such a buffer. The bicarbonate concentration is a little under 2 mmol/kg and the carbonate concentration is a tenth of that. The sort of buffer that one would use in the lab would have at least 10 times the bicarbonate and 100 times the carbonate. So although seawater is buffered, it is very poorly buffered.
Carbonate sediments add greatly to the buffering capacity of the ocean, which is why high CO2 levels in the distant past did not produce acidic oceans. But that is only effective on very long time scales – at least thousands of years, if not much longer. So that will be of no use in buffering the ocean against a big CO2 increase on a time scale of a century.
Similarly, carbon dioxide is a very weak acid even in distilled water.
Normal rainwater pH is below 7 provided that hardness remains below 1. http://pages.uoregon.edu/hof/S01havestingrain/data.html. Freshwater hardness remains usually below 1200 ppm, but tapwater pH goes above 7 also elsewhere http://water.usgs.gov/owq/hardness-alkalinity.html#chart.
Ocean hardness is above 6000 ppm and pH above 8. http://www.tvdsb.ca/uploads/ScienceProbeware/totalwaterhardness.pdf. Sounds like to a powerful enough buffer to me.
I wish alarmists good luck with the project of filling our oceans with normal rainwater.
Mike M
I didn’t seen you make any mention in your analysis of living organisms in the ocean – just the chemistry. As the organisms have a powerful influence on time scales of weeks, not 1000 years, I suspect the buffering is much larger than you infer.
There are many areas of the ocean that are short of one or two minerals so biota are limited by that. Iron is a common deficit. There are several ‘buffers’ that are not able to function now because of a lack of CO2. There is a paper on buffers including one that is still non-functional (several papers actually) focusing on the main ‘El Nino area’ off the coast of South America. The ocean does not act like a beaker of solution. It is filled with life that needs and sequesters CO2.
Personally I doubt the ocean, under any conceivable scenario, could be neutralized. As one commenter said a few days ago, the oceans will not become acidic until the planet runs out of rocks.
Gail, Segalstad and Jaworowski as reference is a weak shot, to say the least…
Seawater is a weak buffer, not a strong one. A 30% increase of CO2 in the atmosphere over 160 years gives a 3% increase of carbon (DIC) in the ocean surface and that gives a pH drop of 0.1 unit.
Hardly measurable with the old equipment, accurate to 0.1 pH unit and zero influence on fish and other creatures in the oceans. The pH drop is mostly near the surface, as the exchange with the deep oceans is very slow. But is measured since 1984 (-0.04 pH unit) with more accurate equipment and calculated from other variables…
Ferdinand, you continually refer to the poor accuracy of old pH equipment, which is only accurate to o.1 pH unit, and is trying to detect 0.1 pH change. Yet the claims of global warming are based mainly on thermometers that are accurate or read out with resolution of 1 degree C or more, and at locations that are several degrees C different from nearby locations. The total global change is less than 0.8 C over the last 150 years. Do you believe the thermometer data set, but not the pH meter set, and why?
Leonard, I have the same doubts of the thermometer trends as of the pH trends, only the pH trends are even worse as that aren’t even fixed places…
I suppose that the satellite measurements are a lot more reliable as these scan most of the surface and don’t have problems like UHI effects. The same for the more recent pH measurements as these are at fixed places and better equipment…
Just curious. What is with the longitudinal sampling artefacts?
Thanks, Aussie. I don’t think they’re artifacts. Oceanographic ships often run north-south transects, both for ease of navigation and to eliminate any variability due to changing longitudes.
w.
Willis,
Thanks for that! As you have spent a far amount of time off-shore I accept that explanation, that now makes sense. I like that from your estimation they would be indicative of oceanographic ships. There is an obvious intention there.
I’m confused, we went from, “a startling data omission that eclipses even the so-called climategate event,” to, “So … was there “pHraud” in not utilizing this data? I say no, there was no fraud.”???
‘we’ did’t go, it was Marita Noon’s blog based on Wallace and Willis E. disagrees to some extent.
Corals evolved in the Cambrian Era with atmospheric CO2 20X higher than at present. The oceans are buffered by alkali rock. Claims that CO2 is significantly lowering pH are utter nonsense.
thus the definition of supersaturated…..
As you add an acid…more buffer comes into solution
Like Gail said…you can’t lower pH until you deplete the buffer…..and that would mean dissolving the entire sea floor..and ignoring the fact that CO2 + plankton/cyano = buffer
CO2 exchanges are mostly in the upper 200 m of the oceans, as that is where most of the fast exchanges with the atmosphere is. Most carbonate rock is deeper and takes more time to dissolve… the pH change of ~0.04 unit is measured at a lot of places sine 1984…
Ferd, you need to learn where that carbonate comes from……..
Corals evolved in the Cambrian Era with atmospheric CO2 20X higher than at present. The oceans are buffered by alkali rock.
Thus Gail’s point about basalt. Though you can’t discard the mantling of the sea floor by both biological and mineral accumulations which would likely run either considerably more neutral or even slightly acidic.
Pathetic sampling for large portions of the oceans. Just like ocean temperature we have no accurate historical data.
Doesn’t phase those positing catastrophe.
Dave
Most geoscience projects have these problems. In the oil and mining industries data is commonly sparsely and irregularly sampled (both in time and space). Statistical methodologies have been developed to deal with these exact issues. Even with regular and well sampled observations the issues similar issues still exist.
The devil is in the details. What assumptions accompany the statistical adjustments, what lines of linear interpolation, what allowances for biological processes, tempurature and ocean circulation?
These can be varied a substantial amount and still meet the objectives of the scientists doing the work. It’s not good data, not good enough to justify the alarm bells.
Dave says…” It’s not good data, not good enough to justify the alarm bells.”
=============================================================
Indeed! Not only is the measured result, even given the best methods, inadequate, but just like in CAGW, the “C” is simply MIA. There are many studies documented at CO2 science which well demonstrate BENEFITS of CO2 in the oceans, and others that show that the theorized harms will not happen!
I note that Ferdinand is consistent in NOT advocating any claimed disastrous harms from the very minor PH changes. I also note the quality of responses to him, and consider this to be a very good example of scientists doing real science, a form of online public peer review.
One of the points Ferdinand makes is that the entire ocean is only in play over long time periods. I would like to suggest that the greater the potential change in the upper 200 meters manifests, the stronger the buffering influence of lower depts. of oceans would occur. n other words, I do not think the buffering affect of the lower depths is linear, just as the greater the increase of surface T manifests, the more lower upwelling o un-warmed water would counter the T increase.
Also, in addition to the many legitimate concerns with any and all attempts (even using the best equipment) to measure the global oceans PH flux, a think not only the date of measurements is critical, but the time of day is critical, as local PH flux daily in many areas is large. Also perhaps poorly considered is decadal changes in ocean currents and disparate ENSO conditions. Those two factors, in conjunction with the massive size of the oceans surface and their DEPTHS, widens the error bars to all the different attempts to measure ocean PH changes.
At any rate, to not even mention the two plus million prior readings, and to intermix modeled results with actual data without clearly noting the difference to political decision makers, is in my view not excusable.
Why? You can only work with what you’ve got. In most areas of engineering and science this is exactly the situation you have to deal with.
Why not you could simply grid them using kriging for each year to achieve a global estimate (alternatively simple declustering method using Veronoi-type methods). Then from each yearly estimate you can construct a time series. Kriging has added bonus of supplying kriging variances as proxy to confidence.
Both suggested methods implicitly deal with data clustering. Of course there are nuances that may need to be addressed (such as choice of kriging method, whether to work in polar or Cartesian domains etc.) but these are always considerations. The temporal issue is not such an issue if you’re modelling a global summary statistic as long as the experimental approach is constant.
[Fixed italics/blockquotes. .mod]