However you do it, you need a max 345 MCM of additional water to cover the shortage of supply to Beirut residents if there’s no improvement to the water network, and if you want them to be completely autonomous.
Correction: Don’t want them to be completely autonomous*
Reducing leakage gets this number down to 207 MCM/yr (approx 40% in best cases). Additionally, there is an option for residents to pump, and there are better days in the year to do so. A flexible utility system can allow for reducing the supply at certain months of the year.
Say 6 wet months means you can be collecting that water and storing it somewhere to be used in the summer, that’s about 50% less supply per year. Shortage reduces to ~100MCM
However you want to spin it, wherever you want to store it, the shortage is still significant and cannot be replenished without budgeting. Storage can be performed underground or on the surface. Surface uses land, subsurface is unreliable. A combination of both is desirable.
Being ideological about technical solutions is detrimental. Since we are far from generating drinking quality water for municipal consumption, and the damage to nature has not occurred it may be worth spending a few million$ on statistics to know exactly how much is needed.
Better yet do the damn census