You could use ion exchange resin to remove the salt, I would use DOWEX-50 in the protonated form. I would wash it several times with water to remove the yellow water soluble impurity which is normally on brand new DOWEX-50- I would mix it with the water at the same time. I would then add to the water IRA-67 (weak base resin) in the free base form. I would have washed it first. This combination will remove the sodium chloride from water.
However if I wanted to see what was in the water, then this desalination treatment with resin would be counterproductive.
If I wanted to find out what organics were in the water, then I would extract with diethyl ether or ethyl acetate. The sodium chloride in the water will reduce the amount of the organic solvent which dissolves in the water. It will also sometimes increase the partioning of the organic solute into the organic phase. I would be then wanting to inject the organic extraxct into a gas chromatography machine after dilution with hexane and drying with sodium sulfate.
If I was to want to isolate metals from the water, then I would be looking to do solvent extraction. Depending on the metal I am interested in I would either use 30 % DEHPA in kerosene or 30 % aliquat 336 in toluene.
The DEHPA (Di- 2-ethylhexyl hydrogen phosphate) will extract many divalent metals from the water, depending on the pH it can extract calcium, strontium, barium and lanthanides. I would strip the metals using nitric acid.
The aliquat 336 is able to extract from sodium chloride solution things like copper, zinc, cadmium, iron and cobalt. I would want to strip (back extract) the metals from the aliquat phase into dilute nitric acid, then I would want to use ICPOES or AAS to measure them.
The general idea is that if you take a clean glass bottle and put in it 200 ml of the water sample, 20 ml of the organic phase. You then seal it and shake it. Then take a 15 ml sample of the organic phase to a smaller vial and add 5 ml of the strip solution. This has the potential to increase the concentration of the metals.
The problem is that if I was to use the DOWEX-50 resin on the orginal water sample then it would remove many of the metals which we want to measure using the solvent extraction pretreatment method.
In general if you want to measure things in salt water, I would try to devise methods of recovering the anaylte without needing to remove the common salt.