Minimum night flow analysis is a clever trick for measuring something you cannot see: the water leaking out of buried pipes. In the small hours almost nobody is using water, so whatever is still flowing into an area is mostly running out through leaks. By measuring that quiet overnight flow, subtracting the little that customers genuinely use at night, and tracking what remains over time, a utility can estimate its leakage and spot new bursts. This guide explains how the method works, how legitimate night use is accounted for, and why fine-interval telemetry is what makes it possible.
Minimum Night Flow in one line: Minimum night flow is the lowest rate of water entering a district metered area during the night, typically in the early morning hours when customer demand falls to its minimum. Because legitimate consumption is small at that time, most of the remaining inflow represents leakage, so subtracting an estimate of legitimate night use from the minimum night flow gives an estimate of the area's leakage. Trending that residual over time reveals when new leaks appear and whether repairs have taken effect.
Water demand follows a daily rhythm: high in the morning as people wake and again in the evening, low overnight when most customers are asleep. During the deepest part of that low, usually somewhere in the window from roughly two to four in the morning, the flow into a metered area reaches its minimum for the day. This is the moment when human consumption is at its smallest, so the inflow at that time is dominated by whatever runs continuously, and the biggest continuous consumer of water in a network is leakage. Cracked joints, corroded pipes, and dripping fittings do not sleep; they flow around the clock.
That is the insight behind minimum night flow analysis. At peak demand, leakage is buried inside a large flow of legitimate use and cannot be separated out. At the nightly minimum, legitimate use shrinks so much that leakage stands out as the main thing still moving. The lower and flatter a DMA's night-time floor, the tighter its pipes; a high or rising floor betrays water escaping underground. Pressure matters too, because leaks flow faster under higher pressure, so night-flow figures are often interpreted alongside the pressure the area was under, and some utilities deliberately reduce pressure at night partly to cut this background loss.
Minimum night flow is not pure leakage, because some real consumption does happen overnight. A share of customers are awake or their appliances run automatically: someone flushes a toilet, an ice machine cycles, an irrigation timer opens, a factory runs a night shift, and a proportion of households have a toilet quietly leaking behind the wall on their own side of the meter. To turn the measured minimum night flow into a leakage estimate, the analyst subtracts an allowance for this legitimate night use, drawn from the number and type of connections in the area and any large night users known to be present.
What remains after that subtraction is the estimate of network leakage running at night, sometimes called the net night flow. Because leaks generally flow at a fairly steady rate through the day while pressure changes, analysts may apply a factor to convert the night-time leakage rate into a daily leakage volume, accounting for the fact that daytime pressure differs from night-time pressure. The exact allowances and conversions are where judgement enters, but the core idea is robust: measure the quiet overnight inflow, remove the small legitimate part, and read the rest as leakage. The trend of that residual over weeks and months is more valuable than any single night's number.
Minimum night flow analysis lives or dies on the quality of the data behind it. To capture the true nightly minimum you need the DMA's inlet flow recorded at short intervals through the night, because a coarse reading can miss the actual low point or smear it into an average. You also need that record kept continuously over long periods, since the whole method depends on comparing tonight's minimum against a baseline and watching for drift. A one-off meter read cannot do this; only a stored, fine-interval time series can, which is precisely what SCADA telemetry produces.
With that data in hand, the analysis becomes a matter of trending. Each night's minimum flow is logged and plotted against the DMA's history, and a step upward flags that a new burst has probably started, often before the water reaches the surface or a customer complains. A gradual creep points to slowly worsening background leakage or a growing number of small failures. After a repair crew fixes a main, the night-flow trend should drop, and that drop is the proof the leak was really the one causing the loss. In this way the graph becomes both the detector and the receipt.
Because a utility may track night flow across many DMAs at once, having every inlet meter report into one hosted platform is what makes the practice scalable. A cloud SCADA system such as Merobix stores each DMA's inflow as a continuous trend and can alarm when a night-flow minimum climbs beyond its normal range, so analysts are steered to the areas that changed rather than manually combing every meter. Ranking DMAs by their night-flow behaviour then directs scarce leak-detection crews to where the loss is worst, turning a laborious survey into a data-led search.
At night, customer demand falls to its lowest point, usually in the early-morning hours, so most of the water still flowing into an area is running out through leaks rather than being used. During the day, leakage is hidden inside the large flow of normal consumption and cannot be separated out. Measuring the quiet overnight minimum lets leakage stand out as the main thing still moving.
Legitimate night use is the real consumption that still occurs overnight, such as toilets being flushed, appliances and ice machines cycling, irrigation timers running, night-shift industry, and customer-side plumbing leaks. This amount is estimated from the number and type of connections in the area and any large night users, then subtracted from the measured minimum night flow so that what remains is an estimate of network leakage.
Capturing the true nightly low requires the inlet flow to be recorded at short intervals, because coarse readings can miss the actual minimum or average it away. The method also depends on comparing each night against a long baseline, so the data must be stored continuously over time. Only a fine-interval, continuously logged time series, which SCADA telemetry provides, supports reliable night-flow analysis.
Merobix reads your field devices into a cloud SCADA - the real thing behind these terms, live in days from any browser.