Detection of untreated sewage discharges to watercourses using machine learning

Abstract

Monitoring and regulating discharges of wastewater pollution in water bodies in England is the duty of the Environment Agency. Identification and reporting of pollution events from wastewater treatment plants is the duty of operators. Nevertheless, in 2018, over 400 sewage pollution incidents in England were reported by the public. We present novel pollution event reporting methodologies to identify likely untreated sewage spills from wastewater treatment plants. Daily effluent flow patterns at two wastewater treatment plants were supplemented by operator-reported incidents of untreated sewage discharges. Using machine learning, known spill events served as training data. The probability of correctly classifying a randomly selected pair of ‘spill’ and ‘no-spill’ effluent patterns was above 96%. Of 7160 days without operator-reported spills, 926 were classified as involving a ‘spill’. The analysis also suggests that both wastewater treatment plants made non-compliant discharges of untreated sewage between 2009 and 2020. This proof-of-principle use of machine learning to detect untreated wastewater discharges can help water companies identify malfunctioning treatment plants and inform agencies of unsatisfactory regulatory oversight. Real-time, open access flow and alarm data and analytical approaches will empower professional and citizen scientific scrutiny of the frequency and impact of untreated wastewater discharges, particularly those unreported by operators.

A Low-Cost Water Depth and Electrical Conductivity Sensor for Detecting Inputs into Urban Stormwater Networks

“…This is because domestic wastewater’s temperature is significantly higher than that of groundwater [ 19 ], and the water conductivity is highly correlated with nitrate, phosphorous, and dissolved metals which usually link to specific industrial activities such as metal plating [ 7 , 10 , 11 ]. Instead of only analysing the absolute level of water level, temperature, and EC, the general patterns or rather the rates of change are also considered as strong indicators of illicit discharges [ 20 , 21 ].…”

Estimating SARS-CoV-2 prevalence from large-scale wastewater surveillance: insights from combined analysis of 44 sites in England

“…Low concentrations could be caused by unusually high per capita flow rates (such as groundwater infiltration), or degradation of RNA during transit due to physical or chemical characteristics of the network (such as numerous pumping stations, or consistently different pH). The method of sample collection, together with limited homogenisation of the 'sewage parcel', could also lead to unrepresentative (either low or high) concentrations, or indeed unaccounted sewage discharge could also affect measurement (Hammond et al 2021;Ort et al, 2010). Including an additional factor to account for degradation might provide a better model assumption for sites showing relatively low concentrations (e.g.…”

Assessment of the pollution incident performance of water and sewerage companies in England

“…Failure to self-report pollution incidents can exacerbate the impact to the water environment. In addition, omissions from WASCs prevent the implementation of effective rapid management responses, investment into failing assets and investigation/prosecution of WASCs [ 5 ]. Variation in the WASC operating environment was shown to influence assessed pollution performance, but differences in ranked performance were limited.…”

“…Results from Hammond et al (2021) support this assertion and indicate that WASCs have historically under-reported pollution incidents. This leads to uncertainty in the frequency, extent and impact of pollution incidents on the water environment, hindering intervention, investment and prosecution [ 5 ]. From reported pollution incidents in 2019, 88% were listed as sewage materials.…”