I thought that originally, reading the second paper. But that of course would not work precisely because of (variable with temperature) evaporation.
It does work.
Evaporation is low, even at 60 deg C. No doubt it has some effect. Of course they can predict the amount with great precision, based on physics and calibrations. No doubt they tweak the IV pump syringe gadget a little to make up for the extra water that goes out in evaporation. And when they do that . . . Whattaya know! It adds in just enough to cover losses from electrolysis and evaporation! So, the numbers from the gadget tell you how much evaporation there is. They tell you there is no recombination, because if there was, there would be too much water. In these conditions, recombination adds back in more water than evaporation removes. Run the numbers and you will see.
But it is a mess, it cannot be used to exclude recombination unless evaporation (which is significant) is quantified.
It is not a mess. It has been understood in detail since 1831. It has been observed countless times. Electrochemists know how to deal with it. Losses from electrochemistry and evaporation can be measured separately with resistance heating calibration. They are easily separated even when they occur at the same time. They are easily quantified. If you would make the effort to run a few numbers, you could quantify it yourself.