It has never been seriously proposed that reservoir-induced seismicity is due to the weight of the impounded water. That weight would add such a tiny increment of stress at nucleation depths that it could not have any effect. The increased seismicity is generally thought to be due to increased pore-pressure due to the very large "head" pressure of the water as it seeps down to fault depth. This may result in lubrication of the fault, or, perhaps, slightly wedging the sides of the fault apart(?).

An additional objection to tidally-related seismicity (in addition to the negligible effect of tides on differential stress), is the fact that, at least with strike-slip quakes, the change in stress is equal on both sides of the fault.

The above is the opinion of an amateur based on quite possibly faulty recollection of previous readings . . .

Michael F. Williams
Arroyo Grande, CA USA