I suspect they used 1V, but they do seem to be going for "pretty pictures" over "hard data" (they didn't even label the sub-graphs (a), (b), (c), and (d)!).
I think it accentuates any resonances they find, by an amount that could be significant (e.g., easily 10x) depending on how "low" their "low impedance power supply" model is.
Thanks for your input John... I think that particular paper can probably be put in the "not so useful" pile.