A utility that reports "days of storage" usually means one number: finished water volume divided by average daily demand. That figure is useful for budgeting, but it is a weak guide to what happens when the source goes quiet. A tank that looks full on paper can deliver far fewer usable days once the draw rate rises, the pressure floor is breached, and the lower part of the tank becomes unavailable to the customers who need it. The operating question is not how much water is stored. It is how the stored volume falls over time, and which customers lose service first as it does.
The threat side of that question is getting more attention. The The Troubling Rise in Water Violence interview by Brett Walton, published by Circle of Blue, frames the risk around conflict zones, where the summary notes that "your water supply could be cut" and that damaged dams can flood downstream communities. The piece is not about any single utility, and it does not tell a municipal operator how much storage to hold. It does make a planning point that applies at home: an interruption can come from a failed source, a damaged conveyance line, or a deliberate act, and the clock that governs the response starts the same way in each case.
Drawdown is a curve
Plot storage level against time under at least three demand cases: average day, peak day, and an emergency case with inflow reduced to zero or to a fraction of normal. The average case produces a gentle slope. The emergency case produces a curve that steepens as demand shifts toward the highest service points and as system losses continue regardless of supply. The bottom of the curve is set by hydraulics, not by the tank's nameplate capacity.
The lowest usable elevation is the level at which the system still holds the required service pressure at its highest customers. Below that line, booster pumps may pull the tank down further, but the water they move is no longer the water the distribution system was designed to receive. Pumps that run dry or draw near the tank floor can pull sediment into the suction side, and that sediment ends up in the mains. A tank with 72 hours of volume on the sheet may have 30 hours of usable supply once the pressure floor is respected. The exact figure depends on the system, which is why the test has to be run on the system itself rather than inferred from a schedule.
Pressure zones decide who loses service first
Multi-zone systems draw down unevenly. A zone fed by a single tank, or by a pressure reducing valve from a higher zone, can run dry while water remains in other tanks across town. Interconnections that look reliable on a map may depend on valves that have not been exercised in years, and some may require manual operation at a site that is hard to reach during an emergency. The inventory has to include the position of each valve in each emergency configuration, along with who is authorized to change it.
Hydraulic models are often run at average demand, because that is what the calibration data covers. That run tells the operator that the system is adequate under normal conditions. It does not show where the pressure floor fails. A model that closes a source, holds demand at peak, and tracks each zone's head will show the failure point, and in most systems the first zone to fail is not the one with the lowest tank level on the dashboard.
Write the test before the event
A usable storage figure needs four inputs that are rarely kept together: current tank levels with confirmed sensor accuracy, the minimum pressure the system is designed to hold, the valve configuration for each emergency scenario, and the demand assumption used to compute the draw. Tank level sensors can read high or fail silently, so a drawdown estimate that starts from a telemetry value should note when the sensor was last checked against a manual reading.
Demand reduction is a real lever in an emergency, but it is a behavioral assumption, and the file should not depend on it. Run the curve with no voluntary reduction, then run it again with a stated reduction, and keep both. If the second run is the only one that meets the target, the target is not met.
Where outside help fits
A supply interruption can also damage treatment capacity or allow contamination into depressurized mains, and restoring either can take more time than restoring flow. Utilities without in-house capacity often bring in local RO specialists to assess skids, membranes, and pretreatment after a disruption. The contract should name the equipment covered, the expected response time, and who certifies the output before it reaches customers. A vendor's availability is a schedule item, not a guarantee, and it belongs in the same file as the storage curve.
What to keep on file
The useful record is not a larger storage number. It is a set of figures that states where the number stops being true: usable days by zone, the minimum pressure used in each run, the valve list with its last exercise date, the names and response times of contractors, and the date each sensor was last verified. A utility that keeps that file can show a board or a regulator exactly which zone fails first, and when. A utility that keeps only the tank count can say that water is stored, but not how long it will reach the people at the end of the line.
Further reading: local RO specialists.