Modern sewage systems are a marvel. Garbage disposals and flushing toilets make is easy to whisk away wastewater, sight unseen in underground pipes. It goes off to wastewater treatment plants to remove contaminants, preventing water pollution from raw sewage. But what happens when sewage systems get blocked? As with most things, there are two types of maintenance, reactive (when equipment fails) and preventative (scheduled cleaning and inspections). Preventative includes monitoring the pipes for obstructions, like tree roots, a buildup of solidified fats, bags, baby wipes, and construction waste. Closed-circuit television footage has been used to monitor for large obstructions but has its limitations associated with costly and time-consuming manual assessments. 

Researchers wanted to identify an alternative to CCTV that could provide predictive blockage modeling, based on customer behavior and the surrounding environment. With South Australian Water Corporation being an early adopter of the Internet of Things technology in wastewater systems, could water-level sensors be the solution? In the paper “Proactive Detection of Wastewater Overflows for Smart Sanitary Sewer Systems: Case Study in South Australia” in the Journal of Water Resources Planning and Management, authors Nhu Cuong Do, Luke Dix, Martin Francis Lambert, and Mark Leslie Stephens outline the challenges with using ultrasonic sensors in sewer systems and explore using data from the sensing network to identify early detection of choke events. Learn more about their research and the opportunities for monitoring the hydraulic behavior associated with wastewater blockages at The abstract is below. 


Sewage systems are built to carry contaminated wastewater from domestic discharge points to collection points for treatment. However, their capacity can be reduced by obstructions, displaced pipe joints, or broken pipes, creating abnormal hydraulic conditions. Wastewater overflows are a potential consequence of these abnormal conditions, which pose a direct threat to the environment and human health. This paper describes a permanent continuous monitoring system of a real sewage network using ultrasonic water level sensors for the purpose of blockage/choke detection, as installed in the suburb of Stonyfell, South Australia. From 62 available data sets collected over 1 year, two distinctive features of growing chokes were identified, including irregular peaks and durations that the water level remains irregularly high in the sewer maintenance holes. An early choke detection method was formulated based on the later feature, which continuously scans the near-real-time data to find time periods containing these abnormal water levels. Application of the methodology showed that the proposed method is effective in detecting possible chokes and overflow events before they occur. In most cases, the warning from the first detection is early enough to allow proactive maintenance attendance to be scheduled by the water utility.

Explore the case study findings in depth in the ASCE Library: