You can now use hand refracto meter to measure directly the salinity of water that would help in to avoid translating data from conductivity which is depended upon various factors. Regards

1)  I don't think anyone worries about this.  According to the Dickinson guide for best practices of OA measurements, you only need to know salinity to the +/- 0.1 units (this pub should be your basic bible for all things related to measuring components of the carbonate chemistry of seawater).  That's well within the error of most instruments that use conductivity to calculate salinity.  See the Fofonoff paper for gory details about the calculation of salinity from conductivity.  I would not recommend a refractometer.  Most of those are not accurate to the 0.1 level

2)  Not much.  Obviously it affects the relative amount of carbonate, bicarbonate, and carbonic acid in the water.  It can also affect the speciation of metal (see Millero paper).  The major constituents will be unaffected. 

Article Algorithms for Computation of Fundamental Properties of Seawater

Article Effect of Ocean Acidification on the Speciation of Metals in Seawater

Article Guide to Best Practices for Ocean CO2 Measurements

1) In a theoretical sense, yes. Since conductivity is a measure of the flow of an electrical current, changing the concentration of ions (such as H+) will change the conductivity.

In a practical sense, no. The change in conductivity due to ocean acidification and increasing H+ concentrations would be insignificant compared to the conductivity of seawater due to all the dissolved salts that are already in the ocean...and therefore there is no need to worry that oceanic salinity measurements would start to be wrong.

Here's my rationale: The concentration of H+ in the oceans is 6.3 x 10-9 mol/L at a pH of 8.2. If there were catastrophic acidification and the pH decreased to 7.0, the concentration of H+ would be 1 x 10-7 mol/L. [Note that I am not suggesting that the ocean's pH would get that low anytime soon, just using it as an example.] Even though that is a ~15x increase in the concentration of H+, the concentration of H+ is still minuscule compared to other ions in seawater. For example, the most abundant cation in the ocean is Na+, which has a concentration of roughly 4.7 x 10-4 mol/L. Note that this is more than 3 orders of magnitude greater than the concentration of H+ in a super-acidified ocean (and doesn't even consider the contributions to conductivity of all the other ions in seawater).

2) I suppose that ocean acidification could, theoretically, change the solubility for certain salts or might increase the chemical weathering/dissolution of certain rock types that would then change the ratio of salts in seawater. As a first guess (literally, a guess), I wouldn't expect either of the things I mentioned to change the ratio of different salts by significant amount given the predicted levels of acidification and the huge volume of the ocean. Maybe a marine geochemist will show up to provide a better answer.

Thanks everyone! I guess then that it could only be a concern working in carbon capture and storage leakage scenarios, when we simulate both (and compare single and combined effects of) very low pH (eg 6) and high salinity (eg 50). And even then, probably not significantly.  Thank you for your helpful advice .

The uptake of CO2 by the oceans does in fact change the absolute salinity of the oceans. The change from pre-industrial to the present is within the uncertainty of the measurements; however, if the increase continues in the business as usual scenario it will eventually become measurable.  Conductivity will be able to account for some of the changes but not all, based on the limited measurements available.

For most practical purposes the changes would be too small to be significant, but in some cases, such as ocean circulation, they may eventually need to be considered, but not until far into the future.

The calculations can be found in the attached paper.

Article Estimating absolute salinity (SA) in the world[U+05F3]s ocea...

Thank you Ryan, that was really helpful.

I think the best and the sure way is to measure salinity directly using a salinometer . this may avoid any translating measures from conductivity. Nevertheless, to know the global physico-chemical water composition may help also in understanding any variation observed in salinity values.

Mohamed, a salinometer uses conductivity to measure salinity. There is no direct way to measure salinity. The most precise and accurate method is through conductivity.

Thanks Ryan . I mean that with a salinometer we can get directely  a salinity value taking in mind that the translation (from conductivity) was done "automatically" .

It's a good question!  I would recommend this paper: 

https://www.researchgate.net/publication/279892497_EFFECT_OF_THE_NON-IDEAL_COMPOSITION_OF_SEA_WATER_ON_SALINITY_AND_DENSITY

In a few words, error in salinity estimation caused by CO2 change is dS=0.03*mmol/kg, so historical change CO2 in 100umol/kg (which is quite common) could cause error in salinity estimation comparable with an accuracy of a precise salinometer.  

In 1960-70th (before the PSS-78 introduction) a lot of studies were conducted to establish S(C,T,P) relations with counting other influencing parameters - alkalinity, CO2, silica (Brewer, Park, Bradshaw, Connors,Weyl, Cox, Culkin, Rilley). Nowadays researches just take the PSS-78 as granted, or, in a best scenario, they are simply refer to the TEOS-10 absolute salinity definition as  a "true salinity". But TEOS-10 absolute salinity is still depend on uncertainty of estimation of the practical salinity. Even the best accurate salinometer Autosal gives methodical error in estimation of practical salinity of deep water sample in a much bigger order than an instrumental accuracy of the Autosal itself. For example, reduction in O2 during warming of samples in a lab and measuring samples at different to in situ temperature could source error in salinity (in my estimation) around 0.003-0.005 even for Atlantic deep waters, where composition of sea water very close to composition of the standard seawater (taken from the Bay of Biscay). I published my research and method how to conduct salinometry in conditions of biogeochemical changing of sea water composition: https://www.researchgate.net/publication/234247124_A_New_Method_of_Salinometry , but I still didn't find much interest in the oceanographic community to test  this method in other areas of the World Ocean, where this effect could be more significant: http://salinometry.com/thermodynamic-salinometry/

Article A New Method of Salinometry

Article EFFECT OF THE NON-IDEAL COMPOSITION OF SEA WATER ON SALINITY...

Very interesting question and answers...

Chip-based salinity measures can give us also impedance ("conductance" at different frequencies) which results in higher sensitivities than those at sub-MHz frequencies.

Gold finger electrodes are used to measure the impedance in the MHz frequency range. Impedance changes with concentration and the relative importance of each ion (H+, Na+, Cl- etc) can, theoretically, change with frequency.

But, please let me tell you that I am not an oceanographer but an electrophysiologist...the principles should be, however, the same.

See this paper for more: https://www.hindawi.com/journals/ijocean/2013/529674/