Plugging the data gap to enhance standardization in monitoring combined sewer overflows
Combined Sewer Overflows (CSOs) have long posed significant challenges to environmental water quality in the UK (Department for Environment, Food, & Rural Affairs, Defra, 2020). Designed as safety valves within sewer systems, CSOs discharge excess stormwater and sewage into water bodies during heavy rainfall to prevent urban flooding and avoid overloading wastewater treatment works (Ofwat, 2019). However, the increasing frequency of extreme weather events, urbanization and ageing infrastructure have amplified the frequency and environmental impact of these overflows (Environment Agency, 2021a; House of Commons Environmental Audit Committee, 2022). Pollution events associated with CSOs can degrade aquatic ecosystems, and the sector's seeming inability to reduce their occurrence has eroded public trust in water utilities (Water UK, 2018).
The data dilemma
Central to addressing CSO-sourced pollution is the effective monitoring and interpretation of Event Duration Monitor (EDM) data (CIWEM, 2017). EDMs record the frequency and duration of CSO dis- charges, providing information for regulatory compliance and system optimisation. Recently, the water industry in the United Kingdom has moved towards installing EDMs on all CSOs. However, a lack of data standardization and inconsistencies in data interpretation have emerged as significant barriers to progress (Chartered Institution of Water and Environmental Management, CIWEM, 2023; Giakoumis & Voulvoulis, 2023). These discrepancies can arise from variations in monitoring equipment, differences in data collection and processing methodologies, and discrepancies in how data are reported and shared among stakeholders.
Another well-known yet recalcitrant issue is the prevalence of poor data quality from EDMs. These inaccuracies may result from sensor malfunctions, data transmission errors or environmental factors such as debris interfering with the EDM and can misrepresent the actual number of spill events, leading to misguided prioritization and resource allocation (Garofalo et al., 2017). Improving the confidence in EDM data collection and analysis is paramount, as the accuracy of spill events impacts effectiveness of management and also credibility of reporting (PricewaterhouseCoopers, PwC, 2023).
Challenges of resource allocation
Water companies face challenges in managing increasing data volumes and maintaining sewer networks (House of Commons Environ- mental Audit Committee, 2022). As sewer intelligence improves, maintenance needs will become clearer, though costs may exceed cur- rent practices (Garofalo et al., 2017). Shifting from reactive to preventative strategies offers long-term benefits but requires significant investment in technology, workforce expansion and training (Environment Agency, 2021a; Ofwat, 2019). Balancing these financial demands with regulatory requirements and customer affordability is difficult. Recruiting skilled personnel in data analytics, environmental science and engineering also remains a challenge due to industry com- petition (Environment Agency, 2021a).
The need for standardisation
The absence of a unified approach to CSO data management across the UK's water industry hinders progress in several ways (CIWEM, 2023; Giakoumis & Voulvoulis, 2023). Without standardization, comparing performance and sharing best practices becomes problematic. There is a strong case for industry-wide collaboration to establish standard protocols for data collection and interpretation (CIWEM, 2017; PwC, 2023). Standardizing these methods would facilitate more accurate benchmarking, a culture of transparency and enable regulators and stakeholders to make more informed decisions, towards better financial and environmental outcomes (Water UK, 2018). Moreover, standardization can streamline regulatory compliance by providing clear guidelines for reporting and accountability. Of particular importance in this context is the ‘cleaning’ of spills data. As the evidence available is often insufficient to enable companies to distinguish a ‘real’ event from a false one, or indeed to identify its precise cause, simply by looking at the data. Consequently, data cleaning becomes highly variable with different assumptions, rules and algorithms being used to remove false data. This is not to say that there is no equivalence in the accuracy or quality of storm overflow data reported by different water companies, but unless there is a common approach to data cleaning, such equivalence can only be presumed, not proven.
Technological innovations
The application of artificial intelligence (AI) and machine learning can enhance data processing capabilities (Garofalo et al., 2017). AI can help identify patterns and anomalies in large datasets, reducing the burden on human analysts and increasing the accuracy of spill event detection. Predictive models can forecast potential overflow events based on weather patterns, system load and historical data, allowing for pre-emptive actions. Additionally, low-cost telemetry and sensor technologies can expand monitoring coverage, providing real-time data that supports proactive maintenance strategies (CIWEM, 2023). Implementing Internet of Things (IoT) devices within the sewer network enables continuous monitoring of system performance. A distributed real-time approach for mitigating CSO and flooding in urban drainage systems has shown promise in enhancing system responsiveness (Garofalo et al., 2017). However, integrating these technologies requires careful planning to ensure compatibility with existing infrastructure and to address cybersecurity concerns.
Plugging the data gap
EDM data quality remains a significant challenge. Water companies acknowledge encountering various data quality issues linked to obstructions in sensors, inconsistent communication links, data corruption and poor calibration. These problems can lead to inaccuracies such as false positives or negatives in spill event reporting, complicating efforts to monitor and manage CSOs effectively. While companies assert in their Pollution Incident Reduction Plans (PIRPs) and other documents that they have processes to address these data quality issues, there is a noticeable lack of transparency regarding the specific actions taken. Publicly available information seldom details the techniques or algorithms used to clean and validate EDM data. References are often made to the adoption of ‘Enterprise Data Architectures’ or proprietary data management platforms but without further elaboration. Different architectures will inevitably result in inconsistencies in EDM data processing and the level of the playing field with respect to CSO regulatory reporting.
Ofwat's Open Data initiative aimed to enhance transparency by making EDM data accessible through company Application Programming Interface (APIs). However, these datasets and accompanying documentation typically do not outline the methodologies employed to clean and process poor-quality data. A report by PwC on the Open Data initiative highlights these gaps but does not provide details on data cleaning techniques (PwC, 2023). This omission makes it difficult for regulators, researchers and the public to assess the reliability and comparability of the data provided by different water companies. There is already a guidance document under the Monitoring Certification Scheme (MCERTS) titled ‘Requirements for Installing and Using Event Duration Monitors’ provided by the Environment Agency. However, this guidance stops short of prescribing sector-wide data quality or data cleaning
approaches although it does require that management systems define maximum acceptable data treatment and telemetry errors and that data treatment and telemetry verification measurements are recorded and analysed.
While these requirements compel operators to have procedures in place for data treatment and to define acceptable error margins, they do not enforce a standardized methodology across the industry for data cleaning and quality assurance. This lack of prescriptive guidance means that water companies may employ varying assumptions, rules and algorithms to remove false data, leading to inconsistencies in how EDM data are processed and reported. Looking ahead, regulatory changes are on the horizon that may influence how water companies handle EDM data. From 1 April 2025, all new and replacement EDMs at wastewater treatment plants must be certified under the Environment Agency's MCERTS (Environment Agency, 2021b; Future Water Association, 2023). MCERTS sets stringent standards for monitoring equipment, personnel and organizations, requiring a mathematical specification that demonstrates the use of appropriate algorithms and arithmetic. Companies will be required to provide appraisals of these algorithms with sample calculations to prove their robustness across the full range of data. Furthermore, they must justify any data manipulations, such as the use of smoothing or filtering techniques. A recent article by the Future Water Association dis- cusses some of the implications of these upcoming changes and how they may impact water companies (Future Water Association, 2023). While MCERTS certification should be welcomed and will compel utilities to justify their data processing approaches, it still does not man- date a standardized methodology across the industry. This situation could perpetuate inconsistencies in how data are cleaned and interpreted posing challenges for regulators and stakeholders aiming to benchmark performance and make informed decisions based on EDM data.
Recommendations and conclusions
To advance towards more sustainable management of CSOs and reduce pollution events, the following recommendations emerge:
- Establish a national framework for EDM data collection and interpretation to ensure consistency and transparency across all water utilities.
- Allocate funding for advanced monitoring technologies and expand the workforce to handle increased data and maintenance demands. Government incentives or public-private partnerships could alleviate financial burdens and promote innovation.
- Utilize AI and machine learning to improve data accuracy and enable proactive maintenance, reducing the likelihood of pollution events.
- Develop partnerships between water companies, regulators, local authorities and international bodies to share best practices and drive innovation. Regular industry conferences and collaborative research projects can facilitate knowledge exchange.
Addressing the complex challenges of CSOs and pollution events requires a concerted effort from all stakeholders. By focusing on data standardization, embracing technological innovations and engaging communities, the United Kingdom can make significant strides in protecting its water environments. Overcoming resource allocation challenges and investing in workforce development seen as priorities for success. As climate change intensifies and urban populations grow, a coordinated and proactive approach is imperative to mitigate CSO impacts and safeguard the nation's waterways for future generations.
Francis Hassard, Marc Pidou, Paul Jeffrey
References
Chartered Institution of Water and Environmental Management (CIWEM). (2023) Surface water management action plan. London: CIWEM Available at: https://www.ciwem.org/assets/pdf/Policy/Reports/SWM% 20review%20-%20summary%20(1).pdf
CIWEM. (2017) Event Duration Monitoring (EDM) good practice guide. Available at: https://www.ciwem.org/assets/pdf/Special%20Interest% 20Groups/Urban%20Drainage%20Group/EDM-Good-Practice-Guide. pdf (Accessed: 30 September 2024).
Department for Environment, Food & Rural Affairs (Defra). 2020 Storm overflows evidence project: final report. London: Defra. Available at: https://www.gov.uk/government/publications/storm-overflows- evidence-project
Environment Agency. 2021a River Basin Management Plans 2021 challenges and choices consultation summary report. Available at: https://
assets.publishing.service.gov.uk/media/60084326e90e073ed196dd1a/ Challenges_and_Choices_consultation_summary_reponse_210125.pdf# page=104.46
Environment Agency. 2021b MCERTS: requirements for installing and using event duration monitors. Available at: https://www.gov.uk/ government/publications/mcerts-requirements-for-installing-and- using-event-duration-monitors
Environment Agency. n.d. Monitoring Emissions to Air, Land and Water (MCERTS). Collection. Available at: https://www.gov.uk/government/ collections/monitoring-emissions-to-air-land-and-water-mcerts
Future Water Association. 2023 Water company fines, Ofwat, the King and MCERTS. Available at: https://www.futurewaterassociation.com/ water-company-fines-ofwat-the-king-and-mcerts/
Garofalo, G., Giordano, A., Piro, P., Spezzano, G. & Vinci, A. (2017) A dis- tributed real-time approach for mitigating CSO and flooding in urban drainage systems. Journal of Network and Computer Applications, 78, 30–42. Available from: https://doi.org/10.1016/j.jnca.2016.10.014
Giakoumis, T. & Voulvoulis, N. (2023) Combined sewer overflows: relating event duration monitoring data to wastewater systems' capacity in England. Environmental Science: Water Research & Technology, 9(3), 707–722. Available from: https://doi.org/10.1039/D2EW00791D
House of Commons Environmental Audit Committee. (2022) Water quality in rivers: fourth report of session 2021–22. London: UK Parliament Available at: https://committees.parliament.uk/work/891/water- quality-in-rivers/publications/
Ofwat. (2019) Time to act, together: Ofwat's strategy. Birmingham: Office of Water Services Available at: https://www.ofwat.gov.uk/wp-content/ uploads/2019/10/Time-to-act-together-Ofwats-strategy-1.pdf
PricewaterhouseCoopers (PwC). 2023 Open data assessment report exec- utive summary. Available at: https://www.ofwat.gov.uk/wp-content/ uploads/2023/06/PwC-Open-Data-Assessment-Report-Executive- Summary.pdf
Water UK. (2018) A manifesto for water. London: Water UK Available at: https://www.water.org.uk/wp-content/uploads/2018/11/A- Manifesto-for-Water.pdf
Access journal here: https://onlinelibrary.wiley.com/toc/17476593/2024/
CIWEM members can access the new issue of WEJ through MyCIWEM portal.