Dry Weather Flow: What a Dry Day Spill Tells You About Sewer Capacity

Dry weather flow is the average daily volume reaching a wastewater treatment works when it has not been raining, and it is the number every foul drainage decision in a constrained catchment eventually rests on. When a storm overflow discharges on a day with no rain, it is not primarily an environmental story. It is a capacity statement: the network is already carrying more than the flow its permit was sized around, before a new development adds anything.

That matters more this year than last, because the two headline measures of sewer performance have begun moving in opposite directions.

Water companies logged 7,280 suspected dry-weather spills across England and Wales between January and May 2026, about 48 a day, against 23.5 a day across the whole of 2025.

The figures were obtained under freedom of information by Robert Forrester, a former Environment Agency officer, and reported by the Guardian on 1 September 2026. They arrived in the same year the national storm overflow spill count fell by more than a third. Both things are true, and understanding why is what turns a news story into usable evidence for a foul drainage strategy for planning.

What is dry weather flow, and how is it calculated?

Dry weather flow (DWF) is the average daily flow arriving at a treatment works during a period without rain, in litres per day. The Environment Agency sets it as a permit limit, and calculates it from three components rather than measuring it as a single quantity. One of those terms is where almost all the trouble sits.

The same infiltration term reappears, magnified, in the flow a works must treat before it may spill at all. Under the Environment Agency's permitting guidance for storm overflows, flow to full treatment is normally 3PG + I(max) + 3E. That is three times dry weather flow, plus the maximum infiltration the works was designed to absorb. On the network itself, the flow an overflow must keep passing downstream throughout a spill is set by Formula A: DWF + 1360P + 2E.

Read the two together and the point is plain: infiltration is not a footnote to the capacity calculation but a line item inside it, fixed at a number somebody chose years ago, on the assumption that it would stay there.

Compliance with a dry weather flow limit is assessed on the Q80, the 20th percentile of a year's daily total volumes. With 365 daily readings, the Q80 is the 73rd value.

That percentile method is why a works can hold a clean DWF compliance record and still discharge on dry days. DWF is an average condition tested across a year; a dry day spill is a discrete event. One does not catch the other, which is why the spill record is worth reading separately.

What counts as a dry day spill?

A dry day spill is a storm overflow discharge recorded when there has been no meaningful rainfall to justify it. Storm overflows exist as relief valves for wet weather, so an activation in dry conditions points to something other than rain having filled the pipe. The Environment Agency's threshold for what counts as dry is specific, and narrower than most coverage suggests.

Companies have only reported these separately since 2024, so 2025 was the first full year of comparable data. A recorded dry day spill is not automatically an offence, and treating it as one is the fastest way to lose credibility with a drainage consultee. Test the innocent explanations first.

• Catchment drain-down. On a very large catchment, rain falling well upstream can take more than 24 hours to reach the overflow, so the discharge is genuinely storm-derived even though the local gauge was dry.
• Monitor error. Event duration monitors register activations for reasons unconnected to flow. Anglian Water has attributed false readings to signal interruptions and, memorably, a spider's web across a sensor.
• Groundwater infiltration. Where the water table is high, groundwater enters the network continuously and surcharges it without a drop of rain. This is the explanation that matters for capacity.
• Blockage or pump failure. A mechanical fault backs the system up locally, and a single event is a maintenance question rather than a capacity one. A repeated pattern is neither.

Water UK called the 2026 figures unverified, and said it was too early to describe them as dry day spills. That is a fair caution about any single record, and a much weaker answer to a sustained pattern of them at one works across a season.

A single dry day spill is a data point. A repeated dry-weather discharge pattern at one works, across a season, is a hydraulic finding.

Why did dry day spills rise while headline spill counts fell?

Because the two metrics measure different things, and only one is weather-dependent. In 2025 the Environment Agency's event duration monitoring annual returns recorded 291,492 storm overflow spills, a 35% fall on 2024. Total duration was down 48%, and the average overflow spilled 20.5 times against 31.8. The Agency put that improvement down largely to unusually dry weather following a very wet 2024, not to the network having grown.

A dry year suppresses rainfall-driven spills by definition, but it does nothing to suppress spills that were never rainfall-driven in the first place, and where groundwater is the cause only a falling water table helps. So the headline count fell while the dry-weather count climbed. The divergence is the signal. It is the same paradox that makes drought a driver of flood risk rather than a relief from it.

South West Water is the clearest illustration of what that looks like on the ground.

In the third week of August 2026 South West Water discharged for roughly 300 hours, across 664 separate spills at 184 sites, on 16.5mm of rainfall.

Sixteen millimetres spread over a week is not a storm, and no drain-down explanation stretches that far.

The Thames Water crisis supplies the worked planning example. Cranleigh sewage treatment works, serving Alfold in Surrey, has 29% more installed capacity than it requires, 157.6 l/s against 122.3 l/s, and still recorded 1,436 spill hours in 2025. Headroom at the works did not help, because the works was not the constraint. The network feeding it was, and the symptom on the ground was sewer flooding.

Three things are called infiltration, and only one of them is this one

Three different things in UK drainage practice go by the name infiltration, and only the third has anything to do with sewer capacity. Two of them are site tests you commission, to different standards, for different systems. The third describes the public network itself, and no site test will ever detect it. This is where drainage submissions go wrong, and the first two get confused all the time.

• Surface water infiltration, tested to BRE Digest 365. How readily the ground accepts rainwater from a soakaway. A BRE 365 infiltration test supports a surface water design and says nothing about foul capacity.
• Foul percolation, tested to BS 6297. Whether a drainage field can disperse effluent from an off-mains system regulated under the general binding rules. BS 6297 percolation testing is a different test, to a different standard, for a different purpose, and submitting one in place of the other is a common and costly error.
• Groundwater infiltration into sewers. Groundwater leaking into the foul or combined sewer network through defective pipework. Nobody tests for it on your site. It is a property of the network between you and the works, and it decides whether there is room for your flow.

No site investigation you can commission will detect groundwater infiltration in the network between your site and the treatment works.

That third one is invisible from the site: you cannot survey it, design it out, or change it with any amount of on-site work. The only way to see it at all is in the operational record of the catchment, which is what makes the dry spill data useful to a drainage consultant rather than merely alarming to everyone else.

What RPS 362 means until 2030

On 19 January 2026 the Environment Agency published regulatory position statement RPS 362, Discharges from groundwater surcharged sewers. It does not change the law. What it does is state that the Agency will not normally take enforcement action against unpermitted discharges caused by groundwater infiltration, provided the company meets a defined set of conditions, and it runs until it expires on 31 March 2030.

For anyone assessing sewer capacity in England it is the most consequential document in the field, and almost entirely absent from a public debate still arguing about whether the spills happened at all.

1. Notify each discharge, to the Environment Agency incident hotline, before it happens where possible or within 24 hours afterwards, referencing the company's Infiltration Reduction Plan.
2. Hold an Infiltration Reduction Plan that quantifies the infiltration pressures, locates them, and sets out timescales and specific milestones to resolve them.
3. Publish it in a draft drainage and wastewater management plan by November 2027, with the final plan due August 2028 and the Agency's written agreement required before then.
4. Deliver the AMP8 measures before 31 March 2030, applying best technical knowledge not entailing excessive cost, and report progress quarterly from November 2027.
5. Keep the records for ten years, and stop the activity entirely if the conditions cannot be met.

Category 1 and 2 pollution incidents are excluded, and the position does not apply where a discharge is likely to cause significant pollution or harm to human health. From 1 April 2030 the Agency returns to its ordinary enforcement policy.

The regulator has formally stood down, until 31 March 2030, on the precise failure mode that most often constrains new connections.

Read that as a planning fact rather than a scandal. No enforcement action at a works tells you nothing about whether it has an infiltration problem. And the pressure to fix that problem inside the life of your permission is weaker than most developers assume. The IRP, though, is a dated and milestoned document you are entitled to ask about. That is more use than any spill count.

What this changes in a foul drainage strategy

Not the method, but the evidence. A capacity enquiry to the sewerage undertaker is still the starting point. Section 106 of the Water Industry Act 1991 still gives a right to connect that settles the pipe rather than the treatment. What changes is what you interrogate when the answer comes back thin.

• Check the dry-weather record for the receiving works, not the national total. The Environment Agency's event duration monitoring returns are published annually, overflow by overflow.
• Ask whether the catchment sits in an Infiltration Reduction Plan. The plan is a public document with dates in it, and the dates are the point.
• Separate surface water properly. Removing it from a combined system remains the largest single reduction available, which is why a surface water drainage strategy often resolves a foul constraint faster than anything aimed at the foul system.
• Derive foul flows honestly. Flows calculated to Building Regulations Part H and the undertaker's own standards survive scrutiny. Optimistic per-dwelling figures do not. Our free drainage calculators will give you a defensible number first.
• Cross-check the records early. A CON29DW drainage and water search and, where groundwater is the suspected driver, groundwater monitoring will often settle the question while it is still cheap to know.

A dated milestone that falls before your occupation trigger is worth something. A milestone in 2029 is not a phasing plan.

None of this makes a constrained catchment unconstrained. It makes the constraint legible, early, and arguable on evidence rather than assertion. That is the difference between a phasing plan agreed before determination and a refusal argued at committee. It also shapes what you must prove when discharging the drainage conditions attached to the permission, and what the undertaker will accept when the pipework comes to be adopted under Section 104.

Enforcement runs on its own track. The joint Environment Agency and Ofwat investigation into more than 2,200 treatment works opened in November 2021 and is still live, the Water (Special Measures) Act 2025 raised the maximum civil penalty to £500,000 in July 2026, and Defra's storm overflows discharge reduction plan still targets no overflow spilling more than ten times a year on average by 2050. None of it adds a litre of headroom to the works serving your site.

The wider picture of how sewage capacity now shapes what gets built sets out where the constraint bites hardest. Our guide to water companies and the planning process explains who you are negotiating with. Where a scheme depends on connecting into a catchment with a dry-weather discharge record, the evidence to settle it is public. Our experienced consultants prepare foul drainage strategies that answer the capacity question with evidence rather than assertion, alongside the wider drainage strategy work a constrained site usually needs. Free scoping, and a fixed quote within the hour.

Frequently asked questions