The problem is that there is a large hump between the intake and the load. It seems that electrons don't care about travelling against gravity (resistance of he wire and current being the only factors indeterming losses) whereas water does (peaks in penstock creates air pockets).
If I understand this correctly:
Humps in the pipe shouldn't be an issue as long as they're below the inlet and the total rise of multiple humps never exceeds the drop at any distance from the inlet (which could create vapor lock). With that criterion met, water flow should start automatically and eventually carry the bubble(s) away. Even if the initial flow is not adequate to carry the bubbles down, the rising pressure as the water descends the hill will increase the solubility of the air at lower parts of the pipe. If all the humps are significantly below the inlet this will eventually dissolve the air into the water and remove it that way.
Regardless:
If the multiple-hump anti-vapor-lock criterion is NOT met - or if the flow isn't adequate to purge the air in a reasonable time, you can always drill a pinhole leak into each high point. (Or install a purge valve. Or an automatic air purge valve {with an internal float} like those used on a solar heating system that must be drained occasionally to avoid freezing.)
If your pinhole leaks (or other purging systems) get clogged and then things somehow get vapor locked, stick a cotter pin into each leak hole (and go to leak holes as a purge mechanism). As with doing this in the "snifter" hole of a hydraulic ram, it should be jiggled around by the water flow and continuously sweep debris and corrosion from the hole.
The bottom line is that I was considering both of your points, that is why I was also asking about changing my existing Harris Pelton Wheel from a 12 volt system to a 48 volt system.
Or you could tap most of the AC from the existing alternator upstream of the diodes and use transformers to step it up to whatever is convenient for shipping it on higher-tension lines to the load. Then transformer it down to something useful once it gets there. Keep the diodes in place and add a local battery to provide an excitation source at the alternator. The regulator will do what is necessary to provide a clean voltage source.
You'll have more loss than if you generated a higher voltage in the first place - but for significant distances you'd want to use 8 KV or so anyhow and you don't want to have that in your alternator.
Hams used to use this hack to get high voltage high wattage for tube transmitters in their automobiles back when alternators were new.
Don't sweat that the frequency is high: That lets you use smaller transformers. (You have to take it into account when you pick the transformer windings, though.) For about 1/3 mile with elevated wiring skin effect won't be an issue at high audio frequencies. At the far end you can transform it down to something sane for charging your main batteries and go from there.
If you're interested in doing it this way I can describe the hacks for picking the correct transformer. Basic idea is that transformers and generators are a match for saturation effects. So find out the lowest frequency "F" where the alternator will generate the correct voltage. For a delta-connected transformer set with ratings based on 60 Hz the voltage rating for the low-voltage side would be 12 * 60/F. Current rating must meet or exceed the genny's maximum current output. (Scale appropriately if you want to connect Y.) Output voltage would also scale in proportion - the ratio stays the same.
Sky wires will need disconnects and surge protectors at each end, just like the power company.
