Three columns, three different questions
The design influent chart for municipal sewage lists three columns side by side: the national statistical mean, the value below which 80% of surveyed plants fall, and a common design range. They look interchangeable and they are not. Each one answers a different question, and using the wrong one changes both the tank volume and the process that follows. The figures come from a survey of 127 wastewater treatment plants and from industry design practice: they are statistics, not a national-standard value.
The statistical mean: what a typical plant sees
The mean column — COD 220 mg/L, BOD₅ 81.6 mg/L, suspended solids 148.5 mg/L, ammonia-N 22.8 mg/L, total nitrogen 30.4 mg/L and total phosphorus 3.7 mg/L — describes the average of the surveyed plants. It is the right figure for describing a typical load, and the wrong figure for sizing one. A plant sized on the mean is underloaded on half of the days it operates and overloaded on the other half, and because a treatment plant does not fail gracefully at the top end, the mean is a description rather than a design input.
One column even moves against the others: the 80th-percentile figure for total phosphorus, 3.3 mg/L, is lower than the mean of 3.7 mg/L, a property of a skewed distribution rather than a typing error.
The 80th percentile: how high a plant realistically goes
The column headed 80% of plants below reports the value under which 80% of the surveyed plants fall: COD 282 mg/L, BOD₅ 105 mg/L, suspended solids 210.8 mg/L, ammonia-N 27.6 mg/L, total nitrogen 36.4 mg/L and total phosphorus 3.3 mg/L. It is a statistical statement about the population of plants, not a guarantee that a new plant will never see more.
Its value in a design conversation is that it separates the ordinary high day from the extraordinary one. A plant that handles the 80th-percentile load without distress has a working margin, and it is a defensible basis to state in a specification. It is not the same as the peak, which is a hydraulic concept handled by the Kz factor. The two columns are also where a common mistake hides: a supplier quoting against the mean and a buyer checking against the percentile are not comparing the same design.
The common design range: what to actually build
The third column — 250–450 mg/L COD, 100–200 mg/L BOD₅, 150–250 mg/L suspended solids, 25–40 mg/L ammonia-N, 30–50 mg/L total nitrogen and 3–5 mg/L total phosphorus — is design practice rather than measurement. It sits at and above the survey figures on purpose, and it is the column most designs are actually held to.
A range is more useful than a point here, because it shows where a decision is being made. A plant at the bottom of the range is sized for a well-separated municipal sewage; one at the top is sized for a stronger influent, with the extra tank volume that implies. Which end applies should be stated in the specification, because a supplier offering against the bottom of the range and one offering against the top are not offering the same plant.
Why there is no national-standard influent value
It is reasonable to expect a national standard to state a typical influent concentration. GB 50014-2021 does not. It covers design principles, effluent standards and process parameters, and it leaves influent quality to the project: the concentration entering a plant depends on the catchment, the water use and the industrial fraction, none of which a national code can fix in advance. The numbers on this chart therefore carry an evidence level lower than the discharge limits. A discharge limit is a table value in a standard; an influent concentration is an operational statistic, and treating the second as if it carried the authority of the first is the error the three-column layout exists to prevent.
Municipal sewage is not the only influent
Everything above describes municipal sewage. Sewage with a significant industrial fraction behaves differently: a food plant, a textile works and a metal-finishing line present different loads and different ratios between them, so the chart is a starting point at best and the specification should be built on sampling of the actual waste stream. Where industrial wastewater is discharged into a municipal sewer rather than treated on site, GB/T 31962-2015 governs the quality that may be discharged to that sewer.
The export case: load per person instead of concentration
Export projects often start from a different basis. Instead of a concentration from a survey, the load is derived per person: 60 g of BOD and 8 g of ammonia-N per person per day. Multiplying by the served population gives the daily mass load directly, and the plant is sized on that mass rather than on a concentration multiplied by a flow. The advantage of the load basis is that it does not depend on an assumed flow, so it survives a change in the per-capita allowance; the advantage of the concentration basis is that it can be checked against a laboratory sample.
Why the load can decide the tank before the flow does
Influent quality is worth this much attention because the organic and nutrient loads often bind before the hydraulic flow does. A plant can be hydraulically correct — the flow passes through it comfortably — and still be undersized, because the biology is sized on the mass of BOD, nitrogen and phosphorus entering each day, not on the volume of water carrying it.
This is most visible where the waste stream is concentrated. A small flow with a high organic strength can demand more aeration and more tank volume than a much larger flow of dilute sewage, and a design that started from the flow alone will have chosen the smaller plant. The selection guide takes the inputs in the order this implies: the load and the flow together, then the process, then the tank, and only then the box that carries it.