Typo, corrected! ou can see it's GW on the simulated 2030 diagram, though. Speaking of which, one circumstance I forgot to note: the nuclear load-following is relatively uneven in their scenario, which they make up for with a higher utilisation of pumped and electronic storage relative to the nuclear phaseout scenario (and they say so explicitely).

_{*Lunatic*, n.
One whose delusions are out of fashion.}