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.
There is a whole wing of regulation that punishes people and businesses of all sizes for misrepresentations related to omissions of material fact. This one seems pretty glaring and material. You usually don’t get the people going on record as to intent as apparently also happened here. Not as fraudulent as a hockey stick, but still possible to at least get charges filed in a different field of endeavor. Leaving large amounts of data out without addressing it prominently and in full should automatically seem reprehensible to a scientist trying to do honest science.
Even preparers of a mutual fund prospectus know you have to highlight important caveats +and prominently+ in order to not be misleading. Using selected data that raises an alarm seems no different than using selected data that exaggerates investment performance. Not to mention the pretense of prediction that accompanies every press release if not every “study”
Willis, I’m sure you have this under control already but I think this is interesting. If we look at the number of measurements by year we see that after year 2000 we essentially have the same low frequency as in the 1930s. There is huge drop off after 1990. I haven’t look at how this affects the different grid cells, I can only assume that the coverage is still the same but redundant measurements in nearby areas are no longer done to the same extent. So possibly, your coverage graph may be sligthly missleading.
here is a plot of the raw data, between pH 0 and 14. As can be seen there is a likely data quality problem in the mid 1990’s that would likely skew any trend to show “acidification”. There is also a similar problem in around 1960.
http://oi60.tinypic.com/21do9qd.jpg
As for a trend, looking at the graph of raw data immediately above, the mark 1 eyeball says there is no significant trend likely to be found in the data. It looks pretty much flat line from 1910 to present. pH starts slightly above 8, and ends slightly above 8, and doesn’t show any significant pattern trend above or below the line.
Yes if you correct for location, season, and sample density you may be able to narrow the spread of the data and find a trend, but there is also a good chance of introducing a spurious trend.
Ferd, the theoretical trend from increased CO2 levels is 0.1 pH unit since 1850, of which 0.04 pH unit since 1984. You have definitely proven that the glass electrode pH measurements are too inaccurate to show any trend in any direction. Thus the answer of Sabine was right: insufficient quality of the data.
Except if you can combine samples at the same place and season over time, but even then…
Ferdinand by your logic the glass thermometers used to create a statistical record of average surface temperatures cannot work either.
However, that is not correct, because statistics can be used to combine many inaccurate samples into 1 sample of greater accuracy.
What I’m seeing that the underlying data has no trend, because the trend = 0, not because the equipment is inaccurate, but because the data shows that acidification is not happening.
Thus, the data shows that the theory that human activity is acidifying the oceans is contradicted by the data.
Fred
Not so. Whilst a stastically meaningful result can be gained where, for instance, there have been seven readings of a thermometer at the same place and the same time, by discounting the outliers and averaging the rest, in reality that doesn’t happen as there is often just one reading. No matter how much you massage it, should it turn out to be inaccurate,, a bad reading is a bad reading .
The same applies to ph readings. If numerous ones were carried out in one place at the same time I would accept you can extract meaningful data from it. However, I suspect these are highly random readings and their value in part, if not in total needs to be queried.
Tonyb
Tony,
It is even worse for the pH readings: the samples are not even taken at the same place and in the same season… That can give differences of up to 1 pH unit from one year to the next, depending of where and when the weight of the samples was in each year. While one is looking for a trend of 0.1 unit over 160 years…
Ferdinand
I can not see how these can give any sort of meaningful data unless some very specific places are selected where the methodology and sample numbers can be shown to be meaningful.
Tonyb
Nicely done Ferdberple (and others). I think your independently developed time series is consistent with my plot which has been featured in Anthony’s first post on this topic, given that your plot covers all records and that particular plot of mine filtered out all records from depths greater than 200m.
My conclusion looking at the data is that the explanation that it is not of sufficient quality to calculate a pH trend is without foundation:
To the degree that the pH readings are “random” as to location and season from year to year, and the underlying data is bound by the central limit theorem, it can be argued that these readings are simply random samples from a population, and will be normally distributed around the mean for that year, and can be treated as such statistically.
While the above is unlikely to be strictly true, because weather, human populations and geography will favor some areas; weather, population and geography themselves are essentially random.
There is a danger that gridding and infilling will simply create false trends where no trends exist, because it will give significance to areas where data is sparse, and remove significance from areas where data is plentiful.
If there is a real trend in the data, it should be somewhat visible to the eye in the raw data without the need to torture the data to gain a confession. I’m not seeing any sort of trend in the raw data.
I do think however that the data shows more than enough data quality to calculate pH trend (or lack of trend), as aside from some obvious data quality problems, the data shows reasonably tight grouping. It isn’t scattered all over the place as it would be if it was low quality and/or high noise.
Thus, my conclusion looking at the data is that the explanation that it is not of sufficient quality is without foundation.
ferdberple.
You do not get it do you! 🙂
The data that do not support or comply with the theory is not good data, or as they say lately is of no sufficient quality.
There is no buts and ifs there……… theory first then data but only if data supports the theory otherwise it becomes junk data….or worse it gets lost and destroyed in the process of the hockey-sticks and cherry-picking .:-)
I think you have already heard and know about the 97% consensus and the 95% certainty on the theory, whatever theory that is….:-)
The latest and improved modern-day climatology. 🙂
Ferdinand has being telling you that for some time now.
cheers
Ferd, as far as I remember from very long ago statistics, you can obtain meaningful results if the sampling is representative for the whole population. In this case there is a huge oversampling in 10% of the areas and near no samples from other areas.
That is as representative as looking for the average height of people on earth where your sample is 50% composed of people from Scandinavia and the rest scattered over the earth and practically none from South America and Asia…
All what one can do is looking at binned data within one small area and the same seasons and look at the individual trend at that area. Eventually combining them by removing the local and seasonal offset.
shows more than enough data quality to calculate pH trend
Sorry, even after throwing out the worst outliers, the whole cloud is between pH 7 and pH 9. Accurate measurements (colorimetric and calculated) since 1984 show a seasonal variation of 0.05 pH unit (Hawaii) up to 0.1 pH unit (Bermuda). Trend in several places: ~0.04 unit.
Are you really sure that the glass electrode data are accurate enough to show such a trend?
I would question that the readings are necessarily random by season. They might be over much of the planet, but both the arctic and Southern oceans can get right cranky as you move into winter. There’s a reason that many field scientists talk about “field season,” and it is all about climate. This may in part explain the sparsish data around Antarctica.
‘two million plus pH samples taken by oceanographers over decades’
Sounds like a lot but lets us look at that again , its over decades so it reality even an yearly average for limited area not be possible and the ocean is two thirds of the planet with some areas having no or little coverage . so the actual amount of coverage for any area is actual not much. And the old trick of ‘smearing’ the data over a large area is hardly one that can be accepted without question.
The problem remains the data is used not because its know to be a good representation of what is being measured , but be because its ‘better than nothing ‘ So in one way not using can have a degree of justification, the trouble is they replaced it with worse model run ‘data’ with all the problems that brings and simply have not made it clear why this real world data was rejected in the first place and given the authors own backgrounds and former publications , the temptation to get the ‘right’ has opposed to a honest result can be seen has one they may well have fallen for.
Once again in climate ‘science’ we see the professionals working at a standard unacceptable for an undergraduate , the rejection of real life related data without explanation of reason. And once again we can ask , is there actually any standards in this most politicised of ‘sciences’
My greatest take from this post and the comments so far is this by our apparent resident Oceanography expert.
“Ferdinand Engelbeen December 31, 2014 at 6:44 am
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)”
So as with the so called “Global Temperatures” we have a POSSIBLE change over 30 years that is minute in comparison to that which the earth experiences on a daily & annual basis and which humans (and probably all sea creatures) cannot even detect.
And just like with the temperature the Warmists want to waste Trillions to offset for a Possible change of 0.12 pH unit over the next 100 years and that is only in the top 200M of the Oceans.
AC, where in the life of WUWT have I ever said that the minute temperature increase over the past 100 years or the 0.1 pH unit decrease will have any negative impact?
Most sea creatures (coccoliths, corals) evolved in much higher CO2 levels and temperatures.
But my stance is that you can’t lump a lot of inaccurate pH data of different places and different seasons together and call the result a “trend” of ocean pH…
As good as you can’t do that with historical CO2 data measured by wet methods in the middle of Paris or forests or temperatures in growing towns…
Here is the raw data for depth 0, with a quick cleanup for the obvious data quality problems. The trendline shows no acidification.
http://oi60.tinypic.com/9s7xvo.jpg
Ferd, look at the data from the same places in the oceans in the same season below Japan:
http://www.data.jma.go.jp/kaiyou/english/oa/tr_pH_WINTER_137E_en.png
How many of the raw data were taken at the same place in the same season?
your result above for japan is a cherry pick. it is statistically insignificant.
the brute force result previous is essentially a random sample. so long as the data is bound by the central limit theorem the results should be “good enough” to at least get us in the ballpark and tell us if there is anything significant to be found.
As Berényi Péter results below confirm.
ferd, look at the difference in measurement error: +/- 0.002/decade (real equipment) vs. +/-0.038 (statistical, equipment at +/- 0.1).
There are a whole lot more Chinese peeing into the rivers that flow into the oceans south of Japan that there used to be. Not only is this causing the oceans to rise (biggest increase in sea levels are in that part of the world), now we have proof they are acidifying the oceans as well. Hopefully the UN will get a global agreement for everyone to stop peeing before it is too late.
have you stopped to think about what the graph actually shows? 3 reading, each almost 700 miles apart on the ocean, with different currents, different rates of evaporation, rainfall and photosynthesis. each with different pH but EXACTLY the same slope on the graph.
And you don’t think to question? Have you ever seen 3 experiments in science, conducted at 3 separate locations, deliver exactly the same results? Almost as though the experimenters were in communication with each other.
and then stop to ask yourself, in ocean pH is such an issue, why does life not die off every day and every year, given the change in pH due to location and season? the measurement is a nonsense, and the fact that three locations deliver the exact same result over 35 years.shows collusion.
you wouldn’t believe this result if three locations showed EXACTLY the same trend for temperature, rainfall, or any other physical measurement. you would think they were cooking the books. yet you accept if for pH to what, something like 1 part in 1 million? because to my eye that appears to be how similar the slopes are. the data is not believable.
Wouldn’t an obvious explanation be fertilisation in coastal waters from run off? Having dived quite a bit at various world locations I can assure you that I see far more algae on reefs today than I did 30 years ago…
If anything, the trendline in the raw data shows that the surface is becoming less acidic, which would explain why the atmosphere is showing more since 1910. As CO2 has been driven out of the oceans into the atmosphere, the ocean pH is increasing, as is the CO2 ppm in the atmosphere.
So, if anything, the raw data is consistent with the theory that the increase in atmospheric CO2 is not due to human burning of fossil fuels. Rather, it represents a loss of CO2 from the oceans, due to a process that is evidently not yet well understood by science.
Sorry, impossible: there is an increase of total carbon measured in all open oceans over time (lots more accurate measurements than pH). That is because the oceans absorb part of the increase in the atmosphere, not the other way out.
If the oceans were a net source, and humans add twice the amount as found in the atmosphere, where does the rest go? into space? Not in vegetation, as that only absorbs some 10% of human emissions (as proven by the oxygen balance).
History shows there is no shortage of scientists that said something was impossible – only to later be proven wrong.
Ferd,
Berényi found an overall trend of -0.002±0.038/decade for pH measurements
The Japanese found a trend for specific places of -0.015 to -0.018±0.003/decade
As far as I can see, the data of the Japanese are widely within the error margin of what Berényi found, but with much better equipment.
But you haven’t answered my question where in nature the sink is that removes halve the human emissions + all the extra CO2 alleged to come out of the oceans?
where in nature the sink is that removes halve the human emissions
=================
algae. the true masters of planet earth, on which all other life depends.
“shows that the surface is becoming less acidic”…which is what you would expect from plankton/cyano when carbon is not as limiting
Latitude, CO2 was never a limiting factor in the oceans, trace elements (iron) are the limiting factor.
During spring in Bermuda including algal blooms, DIC (mainly bicarbonate) drops with 2.5% in seawater. That is all.
….I can’t believe you just said carbon can not be limiting
Latitude, CO2 is one of the limiting factors in the atmosphere, it is not a limiting factor in the wide oceans.
Total carbon (DIC), of which over 90% bicarbonate, the building block for most shells and corals, at Bermuda in winter/spring: 2060 μmol/kg in summer/autumn. After several algal blooms: 2030 μmol/kg. Doesn’t seem that algae are starving from bicarbonate shortage. But the change in DIC by algae has a huge influence on pH and pCO2.
Ferd…the higher the pH the more limiting Carbon from CO2 becomes
Land plants can get all the CO2 they need by opening their stomata. CO2 supply is limiting when the plants have to restrict stomatal opening to limit water loss by evaporation. Not a problem for marine.
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.
(5×5 deg grid boxes, monthly averages, trends computed for grid box/month combinations with at least 15 years of data – 382 such boxes were found)
If only grid boxes with at least 20 years of data are considered (204 items), ocean pH trend is -0.002±0.031/decade, therefore the null result is reasonably robust.
Berényi,
Have you taken into account the season of sampling: that gives a difference of 0.05 to 0.1 pH unit, due to temperature and bio-life variability…
Yes, trends are only calculated for the same month of the year in each grid box.
I have given a try to grid box / month combinations with at least 30 years of data (81 items). The result is +0.003±0.026/decade, therefore the null result is incontrovertible.
As the measured (and theoretical) trend in several places is about 0.013 pH unit per decade, that still is largely within the error bounds of the overall trends.
Maybe an interesting question: can you look at trend of the 20/30 years after 1989 of the grid box where more accurate measurements were done like Hawaii and Bermuda?
Bermuda was/is sampled at two places:
http://bats.bios.edu/bats_location.html
Hawaii deep ocean measurements are done at 22 45’N, 158W, 100 km north of Oahu.
Yes, but when you average multiple samples and at the same time note temperature can you not compare readings at the same site? Did the pH electrode report proton activity or not? Or are you proposing that it is fundamentally impossible to ever measure pH to within 0.1 pH units?
-0.002±0.038/decade
==========
that shows a very close match with the raw data brute force plot. well within the error bounds.
which is in itself interesting because it confirms that for all the gridding and infilling that is done to try and compute “global” averages, alarm bells should ring if the “corrected” result shows significant statistical difference from the raw data.
+1
And a close match with the more accurate equipment in the Japanese and fixed stations used in the past 20/30 years, well within the error bounds…
Ferd and Latitude,
Have a look at the original “trend” of Wallace for the past 30 years: that shows a drop of 0.3 pH units, which is widely outside the error margin per grid found by Berényi Péter.
So can we agree that the compilation by Wallace doesn’t show the real trend (whatever if that is zero or very small)?
I always love it when I see a supposed rate of change that has an error of +/- 1500-1800%. Meaningless.
James Harlock,
Agreed! The problem is that Ferdberple and Latitude still think that that rejects the few series where the accuracy of the measurements is much better, but still think that a “trend” with glass electrode pH measurements has a meaning…
Ferdinand, You don’t know what I think
Latitude, agreed, one can never know what somebody else really thinks, I sometimes don’t know it from myself…
But I like to know your current thoughts about the pH “trend” compiled by Wallace and the general possibility that glass electrode measurements can show a decreasing trend in the order of 0.015 pH unit/decade over the past 30 years…
“I always love it when I see a supposed rate of change that has an error of +/- 1500-1800%.”
Yes. That’s what you get from the Mike Wallace data which Feely is criticised for not using. For the Hawaii data he did use, the trend was 0.0188 +- 0.0022 /decade.
Correctoion, the trend was -0.0188+-0.0022
“1. Period of stabilization of the magnetic activity Sun.
2. Stabilization period earthly pole position.
3. Descending upwelling in the ocean causing a decline in their photosynthesis and warming seawater.
Reducing CO2 solubility in seawater as a result of heating surface ocean waters, reducing the absorption of CO2 in the process of photosynthesis, increase the content of CO2 gas in the surface waters of the sea due to the increased acidity, decrease absorption of CO2 from the atmosphere and gradually increasing the concentration of CO2 in the atmosphere by diffusion of excess CO2 from seawater into the atmosphere to a level of 300 ppmv over the duration of about 20 000 years Earth’s poles traffic loss and the consequent warming ocean waters about 12 ° C and warming the Earth about 12 ° C.”
No evidence for fraud, WHAT!? Receiving $100,000 from the Heinz foundation specifically for this research and omitting data with no explanation, I call that evidence.
http://www.heinzawards.net/recipients/richard-feely
Read that and tell me what it smells like. Smells like fraud to me.
Robert,
Interesting claim in your link:
“…discovered the startling fact that corrosive waters were at acidic levels not predicted by climate change models to occur for decades.”
I wonder if Feely has any evidence to support the claim that the waters were at the acidic levels?
Does he believe it is accurate?
I am sure that the Heinz foundation only supports research that they like, but that doesn’t imply that what Feely, Sabine and a lot of others have done is false. Feely and others did a lot of other ocean research than pH alone.
And as Berényi Péter has shown, the margins of error from the glass electrode measurements are larger than the trend measured/calculated over the past 30 years. Thus Sabine was right to reject all old pH data made by glass electrodes, as these can’t show the (theoretical) trend…
Thus Sabine was right to reject all old pH data made by glass electrodes
===================
by that logic we should also reject glass thermometer readings, which have been used to calculate a 0.6C temperature change over 100 years.
Ferdberple,
If the margins of error are larger than the trend, then it is impossible to conclude anything from the data. That does mean that the historical pH data are worthless to deduce the real (calculated) pH trend. That doesn’t mean that the data are worthless for other interesting items, as they do give some information about pH distribution over latitudes and seasons, which are much larger than the trend.
But you can’t deduce a trend like Wallace has done from these data and then accuse the researchers of “pHraud”…
I don’t know what the margins of error are for temperature readings, but as that are much more data from a lot more places, the overall error may be smaller, although other problems may arise (UHI, number of stations, change in equipment,…).
“EPOCA was the first international research effort on ocean acidification. The 27-member consortium put many other processes likely to be affected under the microscope. These include nutrient uptake, reproduction, acid-base regulation, nitrogen fixation and primary production. Changes in the dynamics of these biological processes are likely to affect marine flora and fauna profoundly.
The four-year long project produced results based on four themes. One of the major findings under the first, ocean chemistry, is that acidification measured in terms of pH drop in the Arctic near Iceland is 50 % faster than average rates in sub-tropical zones in the Pacific.
Project researchers coupled Earth system models (ESMs), global and regional ocean models, and a sediment model to anticipate how acidification will alter ocean biogeochemistry. Climate change has little effect on acidification except in the Arctic, where freshening from ice melt exacerbates the situation. Regional model simulations highlighted the vulnerability of some near-shore regions as a result of river input and organic matter degradation as well as local upwelling systems.”
http://cordis.europa.eu/result/rcn/85717_en.html
“The same growth mechanism of carbon dioxide content atmosphere occurred by the IPCC (1997 Report), between 120-130 thousand. years, which means that no it could cause a man and his civilization. It is therefore necessary to look for another explanation of the reasons temporary increase of atmospheric carbon dioxide, perhaps such as that proposed above. Climate change, changes in carbon dioxide emissions from the ocean are therefore a side effect of changes biological productivity of the ocean. It is controlled by cosmic processes already described by many researchers, including by Milutin Milankovitch. Warsaw, 25 October 2013, amended on February 6, 2014 year. ”
Bogdan Góralski
“Among the relatively large number of phytoplankton species occurs only a few in amounts that indicate their dominance across the whole Baltic Sea. Generally meet the spring diatoms Chaetoceros Achnanthes taeniata and wighamii and dinoflagellates Peridiniella catenata summer cyanobacteria Aphanizomenon flos-aquae and Nodularia spumigena along with various kryptofitami, and in the autumn diatoms Coscinodiscus granii and Thalassiosira baltica and Rhodomonas kryptofita minute. In winter abundance of algae is very low, and the propagation of very weak or non-existent. In the spring, when it is more more sunlight and the water warms up and is it enough In- żo of nutrients, diatom bloom starts and immediately after the dinoflagellates and various small flagellates. In summer, the phytoplankton is dominated by blue-green algae, and in the autumn diatoms again.”
http://ocean.ug.edu.pl/~oceju/CentrumGIS/dane/morze_baltyckie.pdf
What appears to be pHraudulent is the term “ocean acidification”. It implies nasty acids are going to burn us when we go swimming. The oceans are alkaline (pH around 8) yet we going swimming in rivers and dams that may well be slightly acidic.
The largest variety of aquatic animals prefer a range of 6.5-8.0.
Of course Wikipedia has its usual CAGW/CC/CD correctness: “The pH of seawater plays an important role in the ocean’s carbon cycle, and there is evidence of ongoing ocean acidification caused by carbon dioxide emissions.”
Where is the peer reviewed literature that predicts that the oceans will reach a pH<7.0?
Darn, I was slow to trademark Very Good Idea.
Best get working on Very Bad Thing, Very Good Thing, …
😉
The Warmistas have no other interest in the pH of seawater, other than the to be able to predict the future ocean as a hot boiling acidic catastrophe maelstrom. CO2 present in sea water is 99% a bicarbonate, with barely 1% of it present as carbonic acid These strange people are only interested in fooling the population into believing their weird theory of CO2 being out of control.
I keep hearing that glass based pH electrodes are too unreliable to measure pH to better than 0.1 pH units. That is completely incorrect, such inaccuracy is only seen in low conductivity solutions -which is NOT the case for seawater. My lab routinely measures pH to +/- 0.02 pH units with a combination electrodes and quality assurance is easy using calibrated buffer solutions (NIST traceable). The bad measurements seen in the raw data presented above are probably data entry errors and/or human error. Rejectin all ealy data is patentenly absurd.
HTH
Mark,
Have you systematically studied sea water that is in situ v. days to weeks after sampling? The effects of ussing containers of various compositions? Effects of filtered v. unfiltered? Filtered with fine v. coarse filters? Biologically active v. killed biota? Sampling the top mm, 5mm, 10 mm etc intervals? Noting high temperature gradients over such intervals, esp in still waters? Day v. night measurements? And the factorial combinations of all of the above?
Yes, you are claiming an accuracy for the glass electrode, but you cannot use that accuracy for past times when many of the above are known unknowns.
Sea water is moderately high in ionic strength and needs consideration of effects such as Debye and Huckel described. It commonly has suspended matter that can cause insidious instrumental drift. It commonly has biological activity that can do likewise.
The holistic accuracy is far worse than lab accuracy, rendering not only the glass electrode methods unreliable, but also other methods. If you have ever studied sampling theory for earth science systems, you will be aware that sampling the oceans as has been done, involves large errors irrespective of how pH is measured.
(I wrote a draft Masters’ thesis about relevant electrode complications).
same problem we face with glass thermometers. yet we don’t hear scientists crying out that we should scrap GISS and HADCRUT data in favor of the much more accurate RSS and UAH satellite data. scientific hypocrisy. cherry picking from start to finish.
None of the above affect the pH except ageing the sample, the glass electrode rejects all except protons which is why it works in the first place and is used to measure the pH of sewage ! Do your accept that a Nerstian slope is proof of the principle? A dye method is FAR less reliable in the presence of organics.
HTH
Hmm. I’m sure you could get a Masters from talking about lon known effects on glass pH electrodes. But, you chuck a bucket over the side of a ship, and dip a glass pH electrode in it and note the temperature. Simple, repeated thousands of times and perfectly acceptable. Yet you want to contaminate it by storing, filtering, heating etc. and then using a dye? That complications can be introduced by ‘scientists’ is yet another example of ‘post modern science’ IMHO and assume that all the giants on whose shoulders you stand were idiots. Let me throw your complication back at you, for all reasonable variation in ionic strength of a fresh sea water sample how much would pH be altered?
ferdberple,
Your plot of all the raw data (minus outliers) tells the story accurately given the large amount of data points. It is an excellent presentation of the data without any statistical manipulation. Given the large number of data points it is surprising to me that anyone would raise the question of the accuracy of the instruments given the accuracy stated up to 0.1 pH units (which has been questioned as too large). The trend is important.
The accuracy of the instruments is important when great levels of accuracy are being claimed for the results . You want to claim your results are correct to 0.01 then you better have a means to measure to at least that level. To often we see the notion that we can get greater levels of accuracy by ‘throwing enough maths at it ‘ than the means of measurement is capable of doing . And is before we get to error margins , real life has opposed to lab life , how instruments change over time, how the instruments are used etc.
Oddly there are reasons to take real data with a pinch of salt , for various reasons climate ‘science’ is rampant with this issue , however that does not mean you can just use ‘model data ‘ instead , especially when you fail to mention you did this. In this highly politicised area where many of its leading lights have done much to earn mistrust , you can see why such actions send up red flags to others.
Do you know what the central limit theorem is? Google is your friend…
Gents,
I take a bit of issue with the bandying about the glass electrode. While those in school kits are questionable for low alkalinity water (semiconductor rinse water or subcritical+ utility boilers feeding steam turbines), they are decent for environmental work.
In my line of work I use a variety of qualities in pH measurement equipment. When the field instruments are calibrated on a 2 or 3 point scale with quality buffers, the accuracy will be within +\- 0.01 units. Now, if I take a sample of condensate (re-condensed steam after the turbine) and try to stir it in a beaker and measure the pH, I am wasting my time because a generic probe will not stabilize until the sample has absorbed atmospheric gasses overwhelming the few ppb of ammonia until it stabilizes at around a pH of 5.65 and its conductance is about 0.65 mmoh (uS/cm). We measure contaminants in single and double digit ppb’s. At those concentrations of buffering agents (NaOH, phosphates, ammonia or amine’s), and the pressures (700 psi to 4000 psi) and temperatures 300F to 1150F, it takes very little acidic ions to wreck our equipment. We must measures pH correctly day after day, year after year. Back in the 60’s and 70’s, the industry worried about carbonic acid. We can measure this using cation resin separation of basic ions from anions. The resin releases a hydrogen ion for each cat ionic molecule it takes up. The anions in the sample will associate with the H+ to form the conjugate acid. Acids will give water more conductance than bases. This is known as cation conductivity. Today we know that CO2 will combine with the ammonia to form a non-ionic ammonium carbonate which is ejected in air removal equipment. There is also no supporting evidence that it causes damage. Again, this is in an ultra pure water.
in the job prior, we measured “fresh”water. I quote fresh because the water was taken from near the end of the Brazos River west of Houston, Texas. It has received the effluent of municipal and industrial discharges all along its course. We stored it in a 700+ acre reservoir prior to treatment. The pH of this water would swing a full unit of pH in a 24 hour period. Rainfall and changes of wind direction, seasonal biological activity which will take and release chemicals from/into the water would change the pH in shorter time frames.
Remember what pH is, power of the Hydrogen ion. But also remember that it is not just the H+ that contributes to the acidity or basicity of water.
In seawater, a change in a few ppm of the very weak carbonic acid ain’t going to do much. The ocean is an open system of unimaginable volume. i personally find it difficult to imagine a few more ppm of CO2 in a buffer as strong as seawater will have detrimental affects.
The ultra pure water problems do not apply to seawater… But the thing is, the major buffer is the bicarbonate system so C02 will affect pH. By definition pH IS proton activity…
The statistically significant results point towards seawater pH to be 8 and constant – even at river runoff. Or what else would explain the reduced measuring frequency since 1982? An alarming signal detected from the noise would have surely justified more monitoring resources even then.
In my opinion omitting the results is a serious mistake and can be attributed to fraud, ignorance, stupidity and/or incompetence. The more climate scientists reveal, the less likely fraud seems.
I presume that for the last 100 years individuals researchers have been taking seawater from different locations , filtering to remove biological organisms and silt , and checking the pH with careful changes in ambient CO2 partial pressure and temperature . It seems an obvious series of experiments to do and could be carried out with little expense anywhere.
It must be written up in the standard ocean chemistry textbooks .
S what did they find in these artificial controlled experiments , did the results agree with theory?
All,
One of us needs to find region(s) with substantial density of surface pH observations over an extended period of time (as long as possible, of course) and determine the temporal variability over that period. What is the interdecadal variability and long-term trend? My intuition is that Feely et al did not secure the full story by only looking at a short period (20 years). This is really important work that some of us need to do…cliff mass, university of washington
Cliff Mass:
Agreed that perhaps more work on this unique dataset could be useful. The Website refers to many more parameters than pH that were collected on these cruises, including temperature, salinity, and several carbon measures such as pCO2, etc. Some of these can be used to estimate pH theoretically, and thus would be excellent quality control checks. I tried downloading the data with a few of these variables and successfully received the full dataset (2.4 million lines) in about 10 .gz files. However, the files require more post-processing than I am used to, because each set of data is accompanied by metadata on the cruise itself (NODC Cruise ID), date, time, etc. before the rest of the data appears in normal spreadsheet fashion. Also I had only checked off a few variables instead of the total number that would best allow for analysis of both the carbon chemistry and other ancillary variables.
It would be great if you could get a graduate student involved in investigating the database.
http://www.nodc.noaa.gov/OC5/SELECT/dbsearch/sysinfo.html
Cliff
Have you asked John Kennedy at the British met office? He compiles the SST data for Hadley and will possibly know related ph readings which may well have been done by those royal navy and scientific expedition vessels that came up with the best of the SST readings. Many of them are not worth a light I believe, for a variety of reasons, and John knows my opinion on much of the historic sst database ( pre 1955)
However, there are certain areas and certain timescales where the data is relatively robust.
Also, did the Challenger take ph readings? Those are probably worthwhile as the oldest ph readings of integrity.
Tonyb
One of us needs to find region(s) with substantial density of surface pH observations over an extended period of time (as long as possible, of course) and determine the temporal variability over that period.
========
see:
Berényi Péter December 31, 2014 at 8:58 am
I think that the many pH readings taken in the Caspian Sea do not represent what is happening with the oceans because the Caspian Sea is the largest landlocked sea, with no connection to the oceans.
Also, I noticed that the highest sample densities are often near areas with a lot of industrial and economic activity, and water pollution can affect pH readings. pH readings in those areas can change from increase of industrial activity or prosperous population, and change again from enactment of laws that restrict water pollution. If a dataset is created with all grid cells of the oceans having equal weight instead of all samples having equal weight, I suspect that will tell the story of global ocean pH trends.
If a dataset is created with all grid cells of the oceans having equal weight instead of all samples having equal weight, I suspect that will tell the story of global ocean pH trends
=============
see:
Berényi Péter December 31, 2014 at 8:58 am
and
ferdberple December 31, 2014 at 8:29 am
both approaches generated near identical results.
http://oi60.tinypic.com/9s7xvo.jpg