Given that we know there's water about, it seems unlikely that they wouldn't have reacted, though.
Yeah, spectrophotometry -- so, they're only seeing the high energy electronic resonances of atoms in the material, and expressing it as average composition of probable equivalent content. Absolutely nothing about form, indeed it's averaged over a powder sample (soil or drillings?). I don't know about the exact methodology and response, but it should be typical of others (like XRF).
It should be that, in the raw data, there will be a monstrously huge spike corresponding to O (and probably nothing for H, because H doesn't respond to x-rays -- its highest energy level is only 13.6eV, i.e., UV -- so no measurement of water or hydrocarbons from this process), and then spikes (groups of spikes actually, since many elements, especially heavier ones, have multiple excitations) corresponding to all the base elements shown here.
Or for all I know, they might not even have O (and N, C, F, Li, Be, etc. -- light elements with low energy resonances -- most of which are rare or volatile, at least), and they're simply assuming everything is reasonably oxidized (which is, after all, a reasonable assumption, considering they're drilling recognizable rocks and minerals).
Interesting, in the second graph, the four rightmost "elements" are starred. A note about oxidation state perhaps? Fe would be present as both Fe(II) and (III), after all. That's something you can tell absolutely nothing about in this kind of measurement, unfortunately. At best, you'd have to add up all the elements and their average oxidation states, cross this with the oxygen content, and assign an oxygen balance to whatever the most variable elements are. (Such as, you're extremely unlikely to have Ti(III), and you probably won't have Fe(III) at the same time as Cu(I) or Cu(II). And most oxidized alteration products, like Fe(III) and Cu(II), are probably extremely rare on a world that probably* never had the narest hint of an oxygen atmosphere.)
(*But if we can prove that it did, that would be stupendous.)
Disclaimer: I'm only an armchair chemist these days, but I've done quite a bit of amateur inorganic study before, and I retain quite a lot of knowledge about everything from atomic physics to chemistry and electronics.
Tim