Sewer Water Quality Monitoring: Guangzhou Case

A Guangzhou sewer project used unattended water quality monitoring, wireless transmission and GIS to reduce pollution tracing time from 72 to 12 hours.

Project Background

An older urban district in Guangzhou operates more than 50 km of stormwater, wastewater and combined sewer pipelines.

The existing monitoring method relied mainly on quarterly manual manhole sampling, creating three major problems:

• Slow pollution detection: periodic sampling could not capture sudden wastewater changes. The project had previously experienced high-concentration industrial wastewater entering the sewer network, causing a sharp rise in COD at the downstream WWTP influent and affecting treatment operation.
• Safety risks: some monitoring manholes were more than 3 m deep and could contain hazardous gases such as H₂S, making frequent underground sampling risky.
• Fragmented data: the sewer network was divided into three operational zones, with monitoring data stored separately, making cross-area pollution analysis and source tracing difficult.

The project therefore required a continuous, unattended sewer water quality monitoring system that could detect abnormal wastewater and help operators trace its source.

Monitoring Solution

Spectral multiparameter sensors were installed at strategic sewer locations, including:

• downstream of industrial discharge outlets;
• sewer intersections;
• upstream of the wastewater treatment plant influent.

The system monitored COD, ammonia nitrogen, total phosphorus and turbidity to identify abnormal water quality changes.

For underground operation, the monitoring equipment used a waterproof and corrosion-resistant IP68 enclosure, wireless data transmission and fault self-diagnostics. This allowed data to be collected and uploaded without routine manhole entry.

Monitoring results from different sewer areas were transmitted to a centralized intelligent operation and maintenance platform for cross-zone analysis.

For this project, a COD alarm threshold above 500 mg/L was configured. This was a project-specific warning value, not a universal wastewater discharge limit.

When abnormal COD was detected, the system combined monitoring data with the sewer network’s GIS information to identify suspicious pipeline sections and help narrow the pollution source.

The workflow was straightforward:

Online monitoring → abnormal COD alarm → compare upstream/downstream data → GIS analysis → pollution source tracing

For related instruments, see APURE’s multiparameter water quality monitoring solutions and online COD sensors.

Project Results

The project converted periodic manual sampling into an underground water quality sensing network and reported clear operational improvements.

Indicator	Before	After Project
Pollution source tracing time	72 hours	12 hours
Manual underground sampling	Routine manual sampling	About 20 person-visits/month replaced
Labor cost	Baseline	Reduced by 60%
WWTP disturbances from abnormal influent	About 5/year	About 1/year

The most significant improvement was pollution tracing. Instead of relying mainly on manual investigation after an abnormal discharge occurred, operators could use continuous sensor data and GIS information to reduce tracing time from 72 hours to 12 hours.

The unattended system also replaced approximately 20 underground sampling visits per month, reducing labor cost by 60%. According to the project record, no further underground O&M safety incidents were reported after commissioning.

Earlier detection of abnormal wastewater also helped reduce process disturbances at the downstream wastewater treatment plant from approximately five times per year to one time per year.