Would an irrigation pipeline, rather than a canal, have avoided some or all of those problems? Or are there technical reasons why that wouldn’t have worked either?
A pipeline (or a covered channel) reduced evaporative losses; in any society, drinking and bathing needs are a pretty small fraction compared with irrigation, so societies short on freshwater — Israel’s the poster child of good water management in a desert — should take steps to reduce the evaporative losses, which can amount to more than half of the entire water supply pumped into the distribution channel. That pucker pack we use for shipping parcels and for swimming pool covers would be workable, even for something the size of Lake Nasser — assuming someone didn’t find a way to make the plastic explode. The higher the water level, the less has to be used to generate the same amount of electricity (because the distance it falls), which accelerates the rate at which the lake level rises, etc. Once it’s at its peak, irrigate to your heart’s content. :’) Same goes for those high-altitude canals that irrigated southern California into existence.
https://en.wikipedia.org/wiki/Roman_aqueduct
http://www.diggingsonline.com/pages/rese/arts/other/byzaqueduct.htm
http://www.shc.ed.ac.uk/projects/longwalls/Water/LongSystem.htm
http://www.shc.ed.ac.uk/projects/longwalls/WaterSupply.htm