Why dry weather discharges of untreated sewage happen 

Dr Peter Matthews, Past President, Fellow CWEM, CEnv, examines the recent media coverage of dry weather discharges and the complexities of how it is measured.

The issue of dry weather discharges of sewage has been subject of increasing media coverage. In June this year, the BBC published an article which, it said, pointed to 6,000 possible illegal sewage spills in a year.

The BBC used a paradigm that identified dry weather at the point of discharge as being less than 0.25mm rainfall on the preceding four days, resulting in the 6000 such discharges recorded by water companies in 2022. The methodology was contested by the water companies.

Annual reporting and contestable datasets

Every year, England’s water companies are required to submit an annual report to the Environment Agency (EA) summarising their discharges. This enables the EA to investigate potential cases of dry spills and to decide whether it will take any action. The BBC used data behind the 2022 report. In responding to its findings, the water companies argued that the datasets are unverified and contain errors.

Explanations offered were that monitors provided invalid data, or that some outlets have large catchment areas and it can take a few days for any rainfall to drain through their systems. Also, that naturally occurring groundwater can enter sewers, often from private pipes and in dry weather, which can cause overflows (groundwater infiltration is already embedded in the definition of dry weather flow). No-one cited discharges being due to emergency circumstances.

The BBC responded to the large catchment argument by saying that it accounted for drain-down time by only considering a discharge a potential dry spill when there had been four consecutive days in the surrounding area without rain (although this is not the formal definition of dry weather flow).

Anglian Water’s Head of Spill Reduction, Gail Pickles explained why a dry day spill isn’t always a dry day spill in a blog. She stated that:

“Event duration monitors (EDMs) indicate the start and stop times of when a storm overflow may have been active – and I choose that word deliberately, for a good reason.”

She went on to explain the learning curve of understanding incidence of false activation in what are relatively new installations and the impact of various system disturbances (e.g. high speed trains, network signal interruptions or even spiders) on activations. These then need to be interrogated before inclusion in an annual return to the EA.

Reasons for spills

The diversity of reasons for dry weather discharges can be summarised as:

• Excessive localised cloud bursts of rainwater in a catchment, but overflowing elsewhere from a sewer at a point where the weather is not wet; the delay can be hours, even days, in big drainage areas.

• To elaborate this point, if the overflow is due to rising groundwater levels above the sewer level or private drains and sewers, this could cause an overflow well after the rainfall event and last for a long time. Of course, the reverse might happen with leaky sewers during dry weather and that is equally undesirable with the potential for eventual collapse with blockages and emergency overflows. Sewers may deteriorate by fracturing due to soil movement for a variety of meteorological reasons. So, climate change affects groundwater as well and is an issue water companies are urgently aiming to address, for example Wessex Water have a specific infiltration reduction plan.

• Connections of domestic sewage on an incremental basis under S106 of the 1991 Water Industry Act, with a slow upward creep of base flow, so a sewer can overflow in drier conditions. This may be evident during a daily period when flows fluctuate. This can be more of a challenge for management of sewer assets than treatment works, but it is still a risk for works. There needs to be greater clarity in the relationship between individual connections providing a very small increment of financial contribution and the major provision of additional assets as individual connections increase.

• Big swings in flow over 24 hours particularly in small sewer catchments and works.

• Blockages of sewers due to irresponsible disposal of unflushable materials such as sanitary items and cooking wastes into sewers. It is still allowable to install domestic sink disposal units for food wastes even though this causes problems with blockages.

• Changes in trade effluent discharges; likely to become greater as the UK economy promotes more ‘on-shoring’.

• Emergencies which are a fact of life in sewers and treatment works such as power outages or third party damage to sewers.

• Finally, irresponsible management and lack of investment.

This does not take account of false alarms and how these can be minimised and understood in the matrix of compliance data.

Measuring dry weather flow

At the heart of this is the concept of dry weather flow (DWF) and its measurement. The monitoring of spills is different to that of dry weather flow per se. Both are easiest at treatment plants and most difficult on sewers in remote areas. The BBC article does not offer any distinction.

In the case of permits for effluent discharges, the permissible dry weather flow itself is limited and overspill discharges restricted as multiples of that. If the DWF is exceeded, that is a breach in itself, and the overspill impact may be controlled by lowering the multiple for the overspill, though that may be difficult.

In the case of sewer overspills the DWF is a design feature with overspills at defined multiples, but there is no actual DWF permit limit. Monitoring DWF may be difficult and insights may be gained by the actual versus design frequencies of overspills in prescribed weather conditions. This is at the heart of the BBC-identified issues.

Either way, it is clear that the location of the places in which rainfall is measured, i.e. where it is dry needs to be more clearly defined, understood and agreed. The differences of opinion and practice highlighted by the BBC suggest that just as in other areas of integrated monitoring governance, the definitions of flows and monitoring practices need attention. Is DWF or Q80/Q90 for example, a determinand in its absolute own right or, is it a determinand as part of a companies’ quality assessment? In a way, it is a candidate for a balanced score card approach.

DWF has been defined for permits for many years as the average daily sewage flow entering a treatment works over 7 days with rainfall which did not exceed 25mm on any one day (excluding local or Bank holidays, etc) following a period of 7 days, in which rainfall did not exceed 25 mm on any day. Daily total flows are measured on flow meters certified on the EAs Monitoring Certification Scheme (MCERTS). But this is quite challenging, particularly during our recent wet weather.

So the EA offers the alternative measurement of Nonparametric 80% exceeded flow (Q80). The nonparametric 20-percentile value of a time series of measured total daily volume (TDV) data provides a good estimate of DWF. This value is that which exceeded 80% of the recorded daily values. If there are 365 measured values of TDV in a year ranked from the lowest to the highest, the Q80 is the 73rd value.

However, compliance is also assessed with the 90 percentile flow, Q90,which is the flow equalled or exceeded for 10% of the specified term. This is a high flow parameter which, when compared with the Q95 flow provides a measure of the variability – or ‘flashiness’ – of the flow regime (guidance is available here).

This arcane explanation is simplified in the 2010 EA guidance. Where the measured DWF (90 per cent exceeded daily flow) is greater than the permitted DWF (planned 80 per cent exceeded daily flow) then the company is required to investigate the cause and take appropriate action to minimise the risk of further non-compliance. This is applied to the largest treatment works.

As far as the design DWF goes – be it for sewers or sewage treatment works – the EA provides guidance on the contemporary use of a long-established formula which takes account of design population, per capita water consumption, infiltration and trade effluent discharges. So the current drive for water use efficiency will have the long term benefit of a smaller DWF (but which will be of more concentrated quality).

Conclusion

The issues arising from the BBC article are in many ways all part of the debate about operator self-monitoring of discharges. It has elements of self-monitoring , self-reporting and action taken when non conformity is identified.

It is clear that monitor failure cannot be tolerated, but from the debate there has to be clarity about monitor settings. Many of the reasons given are apocryphal and need more evidence.

From 2025, all water companies have committed to publishing near real-time sewage maps for the public to increase transparency. They will use the data from these monitors to create the maps. And there is no doubt that these will be studied intently by interest groups, regulators, politicians and the media. So it seems reasonable that the reasons for any overspill are given, even if it is the best guess that can be offered.

Furthermore, the measurement or assessment of DWF, by whatever means, needs to be better communicated and understood. Failure to comply does not necessarily mean an illegal discharge has been made and compliance with the DWF does not necessarily avoid making an illegal discharge.

This whole topic could be included in a future effluent quality assurance (EQA) approach. The BBC piece arguably strengthens that position.

The execution of a rigorous programme of monitoring, the maintenance of equipment, the reporting and explanation of non-compliance as part of EQA is the way forward. And that whole tranche of information must then feed into investment plans.

Author: Dr. Peter Matthews, Past President, Fellow CWEM, CEnv