http://www.tresamigasllc.com/docs/nclimate2921.pdf
A quick read of it suggests the following limitations, among others; take the time to read it for yourself if you like:
1. It seems predicated on the notion that we can have a system where any shortfall in one part of the country can be easily made up for just by using the excess generation in other parts of the country. This appears to be a rather "blue sky" assumption to make, and one that I personally wouldn't be willing to make.
2. For the numbers quoted, it makes no mention of "nameplate" capacity vs. actual average capacity, which of course makes a huge difference as to the item above.
3. It makes the rather bizarre statement that HVAC lines can't transmit power over long distances, which clearly isn't true. (They can stretch for thousands of miles if need be, although most aren't anywhere near that long.) It may be true, though, that with current technology HVDC lines can transmit power more efficiently and for longer distances than HVAC lines can.
4. If I'm reading it correctly, it admits to not taking short term variability (an hour or less) into account, when such variability can be substantial. I may be reading it wrong, though.
And so on.
So at best this study can be taken as an example of out-of-the-box, non-traditional thinking, and perhaps as an encouragement for doing more of the same. But it hardly qualifies as any kind of proof that what they describe is technologically or economically feasible.