Radiation fluxes during a direct hit of a SEP event would be much higher. I’m not sure how the plants would fare against such radiation (better than people, I know)… though I am keeping an eye out for this research. I know we wouldn’t want people to be above ground during one. Furthermore, if I’m propagating my plants from flowers (and not clonal lines, kept in a shielded lab), I definitely wouldn’t want them out on surface. Cultural Fidelity (low mutation rate) would be key to long term sustainability.
If you look on Page 10 here (it’s the same article as what you linked, but with more formatting… I’d actually already read it), you’ll see that Curiosity got hit by a couple of SEP’s during transit. The proton flux went from 10e-4 to 10e+1, a 100,000x increase during the event. These things typically last days, so it’s a dose worthy of avoiding when possible.
Also, if your plants are on the surface, then eventually someone is going to have to service the equipment or harvest the plants, and during that time they’re going to be exposed. The way NASA is about radiation exposures, even 0.64 mSv/day is going to give the lifetime risk calculators a fit. Since this risk can be eliminated in its entirety simply by moving everything under the dirt, I figure this is something they’ll want to do.
I don’t disagree though that using the sun would make things much more energy efficient. In regard to energy, my contention is simply that direct solar radiation is not very reliable. Especially, when you’ve got random multi-month-long dust storms to fight against. Solar panels aren’t the only way to go though; there’s also Kilopower Nuclear Generators and fuel cells (recombining your propellant supply of methane and O2 to get electron flow) to get you through the solar droughts.
In the long run, although they are more expensive, growth chambers are more reliable than greenhouses. In those first few years or decades, Reliability wont just be King: it’s going to be the God Emperor of Mars.