Anybunny who has been following Judith Curry's display of purposeful ignorance, the Weasel's rather stoat like contempt for her and her fans Sargasso Sea like level of stupidity and Willard Tony's braying knows that just about no one offering their opinion has a clue about how pH is measured in the oceans and the precision and accuracy thereof. Most of the bunnies, of course, at least those who took general chemistry, have some ideas about pH meters, indeed, in one of the later comments at WUWT, Joe, plain old Joe, puts it quite well
pH Meters are often unstable and not always reliable. They are the fussbudgety spoiled brats of scientific instrumentation. As a chemist, it was depressing how difficult it was to get reliable data, for those of us in OSHA who monitored factory operations. The first point is that, the makeup of the solution can radically affect the readings. as there can be surfactants, a large number of ionic species, and organic molecules. So, one would expect that since the ionic and organic constituents of local oceanic waters vary widely from location to location, it will similarly produce variability in pH readings. The second problem is the inherent instability of the pH probes. They can work reliably (perhaps), but then after sitting for any time, not work. It’s in their nature….
There are two methods, one pretty old, the other newish. First the newish one, an ion selective field SeaFET, accurate uncalibrated to 0.01 pH units, with a precision of 0.005. The SeaFET can be left alone to operate and read out later and here is something about a system under use on the California coast. Basically they are comprised of an ion selective membrane (only lets certain ions pass) on top of an FET.
Second, the old one. Like Joe, everybunny knows that pH meters suck. If nothing else, the damn electrodes always dry out because people don't keep them wet and they cost a fortune. However there is a better way and quite an old one. Many chemicals change color when they ionize in solution. In particular weak acids are happy to do this, and there are shelves in chemistry stockrooms full of different indicators. The basic reaction is
HIn = H+ + In-where HIn is a weak acid and In- is the indicator anion. The concention, [H+] is simply given by
Ka = [H+][In-]where Ka is the acid dissociation constant, a weak function of temperature, or, if the bunnies prefer
pH = -log [H+] = log ([In-]/Ka)determining pH then becomes an exercise in determining the color change as a concentration of [In-]. In one deployed version, the precision is + 0.0007 pH unit sand an accuracy of 0.0005 relative to a reference system the principal requirement being holding the temperature in the spectrophotometer cell constant. Indeed, part of the issue with accuracy is that the method really pushes the ability to mix buffers for calibration. This is OLD technology, although there may have been modern refinements, so Eli would not be surprised to see measurements going back to 1950, if not earlier (stay tuned).