Just to be clear, he was talking to
High School students, so perhaps sacrificed a little accuracy for the sake of making a point.
I still think the general idea is sound, even if you don't take it literally on implementation
Just to address that for the sake of that dude, I didn't want it to come across the wrong way.
To address some things further down:
I do remember a specific detail, and I don't know how accurate it is, however, he did explain that (and also prefaced how theoretical it is, even at that time more so I imagine):
- If you wanted to do this right, what you would want to do is desalinate the water and pump it to existing surface basins/reservoirs first (e.g., the Colorado River Basin), and taking great care to match that water to the ecology (e.g., mineral, PH level etc) of the water you are replenishing This way you aren't forcing an entire set of infrastructure to be obsolete, but rather working with existing delivery modes to get the water where it needs to be
- In cases where this isn't feasible, your best bet would be to run some sort of infrastructure (likely pipes) to those in need. In some cases, you might even be able to simply create new reservoirs to feed population centers to create efficient delivery mechanisms. The other thing I remember him mentioning specifically is there are many delivery mechanisms (e.g natural) ones that were no longer in productive service that could be revived also (I wish I had an example for this, but I don't).
The cheapest way to do this (by the dudes estimation) would be to use nuclear power near the coasts, that rely on the salt water for cooling which gives you two advantages:
1. in theory, the water that the plants themselves could produce a by product of clean, usable water[0]
2. The nuclear power plants themselves should be able to produce the power cheaply and ecologically friendly)[1]
3. This would incentive more research into pressurized water systems for nuclear power, which in theory could yield smaller reactors which in turn are even safer as is intrinsic to their design, they may even yield ever smaller sized reactors[2][3]
[0]https://www.sciencedaily.com/releases/2011/05/110512082949.h...
[1]http://sitn.hms.harvard.edu/flash/2016/reconsidering-risks-n...
[2]http://large.stanford.edu/courses/2017/ph241/kraus1/
[3]https://www.nrc.gov/reading-rm/basic-ref/students/for-educat...
Its, in a practical sense, feasible, but I think the takeaway I had back then and even now, thinking about it, was we needed to start this by 2009 to have it be up and going by 2019. Which isn't to say that we shouldn't do something now, but rather, we're still deciding what to do about it and that's crippling