Sewer & Wastewater Monitoring: How It Works, Sensor Types, and I&I Detection
Wastewater & Sewer Monitoring
A technical resource for wastewater operators and asset managers. Covering how sewer monitoring works in live networks, sensor selection, I&I detection strategies, and what separates useful data from noise.
500+
blockages detected for Sydney Water
15min
typical sensor install time
10yr
battery life (at standard levels)
100000+
devices installed globally
The difference between sewer monitoring and wastewater monitoring
The terms are often used interchangeably, but there is a meaningful distinction. Sewer level monitoring refers specifically to depth measurement inside sewer assets. Wastewater monitoring is the broader discipline that encompasses level, flow, pressure, pump station performance, and in some cases water quality parameters. In practice, most programs begin with level monitoring and expand from there.
What sewer monitoring enables
- Overflow prevention – detect surcharge before it becomes an environmental or compliance event
- Blockage detection – rising levels in isolation from wet weather indicate partial or full blockages
- I&I characterization – Rainfall-correlated level behaviour reveals inflow pathways and infiltration zones
- Asset performance visibility – Understand how your network performs under wet and dry weather conditions
- Capital prioritisation – Data-driven evidence to direct rehabilitation and maintenance spend
- Regulatory reporting – Defensible data for compliance submissions and environmental reporting
How sewer level monitoring works in live networks
01 Sensor installation
02 Measurement & transmission
03 Alerting & trending
04 SCADA & platform integration
05 Analysis & decisions
06 Ongoing management
Sensor installation
A radar or float sensor is mounted inside a manhole or chamber. Modern IoT monitoring devices are designed for fast deployment – most sites can be installed without confined space entry, in under 20 minutes, using flexible bracket systems that accommodate a range of chamber geometries. Mounting position relative to the wastewater surface, pipe inlets, and any obstructions within the chamber is critical to data quality.
The concept is straightforward. Reliable execution in live sewer environments is not. Here is how a well-designed monitoring system functions from sensor to decision.
Sensor type comparison for sewer environments
| Sensor Type | Principle | Sewer Suitability | Key Strengths | Key Failure Modes |
|---|---|---|---|---|
| Radar | Emits microwave pulses, measures return signal from wastewater surface | Preferred | Non-contact; handles condensation, turbulence, and deep assets well; performs through surcharge conditions | Beam obstructions (ladders, pipes, brackets); poor mounting geometry; strong lateral reflections |
| Pressure | Infers level from hydrostatic pressure via submerged sensor | Conditional | Simple and low-cost; works in complex geometries; no beam-path concerns | Fouling and port blockage; sensor drift over time; maintenance burden; damage during high-flow events |
| Ultrasonic | Emits acoustic pulses, measures echo return time to surface | Not recommended | Low cost; adequate in stable, controlled environments | Highly susceptible to condensation, temperature variation, turbulence, foam, and debris - all common in live sewers |
| Float | Mechanical contact measures surface level | Legacy/Limited | Simple; low cost; works in calm conditions | High maintenance; mechanical failure risk; unsuitable for debris-laden or surcharge conditions |
Why radar is the preferred technology for most sewer applications
Radar’s non-contact measurement principle makes it fundamentally better suited to the sewer environment than immersion-based alternatives. It is unaffected by the wastewater itself – its density, temperature, or chemical composition – and handles the condensation, turbulence, and rapid level changes that are routine in live networks.
The critical performance variable for radar in sewers is beam management. Radar measures everything within its beam cone, not just the water surface. Pipes crossing chambers, internal ladders, steps, and joints all produce signal interference if they fall within the beam path. The quality of the install, specifically how the sensor is positioned relative to these obstructions, directly determines data quality. Software filtering can reduce noise, but it cannot resolve a fundamentally flawed mounting position.
Integrated vs. separate sensor and logger
Sewer monitoring hardware comes in two main configurations. Separate sensor and data logger – a radar or float sensor is wired to a standalone data logger mounted in or above the pit. This approach gives flexibility for different chamber types and allows sensor replacement without disturbing the logger.
An integrated radar device combines the sensor and data logger into a single unit that mounts directly in the manhole. Integration reduces cabling, simplifies installation, and eliminates the signal path between sensor and logger. For high-volume deployments where installation speed and long-term maintenance cost matter, an integrated device typically offers the best total cost of ownership.
The right choice depends on chamber geometry, installation constraints, and the number of sites to be monitored. Both configurations are in active use across major utility networks globally.
From alarm monitoring to catchment intelligence
Most utilities begin sewer monitoring with overflow prevention as the primary goal: sensors at high-risk sites, thresholds configured, crews dispatched when levels rise. This is a legitimate and important use case. But it leaves significant analytical value on the table.
Every monitored asset sits inside a hydraulic network. When it rains, the network responds as a system. When multiple sub-catchments are monitored simultaneously, the data begins to reveal comparative hydraulic performance – which zones respond aggressively to moderate rainfall, which recover slowly after storms, which are showing steady dry-weather baseline drift.
This shifts the question from “did this site overflow?” to “how is this catchment performing relative to its peers?” and that is a fundamentally more powerful frame for managing I&I.
Why this matters for capital programmes
I&I mitigation is expensive. Lining, rehabilitation, upsizing, root removal all require significant, sustained investment. Utilities must prioritise carefully and defend those decisions to regulators and stakeholders.
Actual source confirmation relies on CCTV, smoke testing, and flow isolation studies which are resource-intensive methods that cannot be applied across an entire network simultaneously. What level monitoring provides is a ranked evidence base, sub-catchments ordered by hydraulic stress, that directs those investigations to where they will deliver the greatest return.
The difference between stating that an area “appears problematic” and demonstrating that it responds 40% more aggressively to equivalent rainfall than comparable catchments is significant when building a business case for regulatory approval or capital funding.
Read more on how sewer level monitoring creates an I&I map.
The five I&I behavioural indicators to monitor
Dry-weather baseline
Gradual increases in dry-weather levels indicate sediment accumulation, partial blockage, or slow infiltration from groundwater
Rise rate during rain
Rapid level response to relatively low rainfall intensity signals high inflow pathways — direct connections or shallow infiltration
Peak depth vs. capacity
Comparing peak levels against pipe size and grade across similar assets identifies disproportionately stressed zones
Post-storm recovery
Slow drawdown after rain indicates downstream restriction, limited hydraulic capacity, or sustained infiltration from saturated ground
Year-on-year trend drift
Comparing seasonal responses over multiple years reveals whether network performance is stable, improving post-rehabilitation, or deteriorating as assets age
Practical steps to build a catchment I&I performance map
01 Deploy in clusters
02 Look at rainfall
03 Analyse recovery curves
04 Track baseline drift
05 Validate rehabilitation outcomes
06 Calibrate hydraulic models
Deploy in clusters, not in isolation
Monitoring a single high-risk site provides overflow protection but no comparative insight. Monitoring across a sub-catchment enables performance benchmarking between zones.
Building an I&I program? Our engineers have supported utilities across the globe in designing monitoring programs that generate defensible I&I evidence. Talk to us about your network.
Deploying sewer monitoring at network scale
At scale, installation time and safety exposure are major cost drivers. A deployment that takes 30 minutes and avoids confined space entry can be repeated hundreds of times with standard field crews. A deployment that requires specialist confined space teams, structural modifications, or half-day site visits quickly becomes expensive and slow to scale.
Modern monitoring devices are designed to address this directly – magnetic or bracket-mounted designs that clip to the frame, flexible mounting configurations for varying chamber diameters and depths, and pre-terminated cabling that eliminates field wiring. These details directly affect the economics of a large deployment.
Most large-scale programs begin with a structured trial, typically 20–50 sensors across a representative mix of site types and catchment characteristics. A well-designed trial validates sensor performance in the specific network conditions, stress-tests installation workflows, confirms connectivity in difficult locations, and generates the performance data needed to justify programme expansion to procurement and finance teams.
The trial phase is also where data quality problems are most efficiently solved. Issues identified across 30 sites are far less costly to address than the same issues discovered across 3,000.
Deep manholes, dense urban environments, and underground assets present real connectivity challenges for cellular-based monitoring. CAT-M1 and NB-IoT networks offer significantly better penetration into challenging locations than standard 4G, but network availability and signal strength vary by location and carrier.
Devices with high-sensitivity antenna configurations, including external antenna support for the most difficult sites, provide coverage options that allow programs to proceed even in locations where standard devices would fail. This matters in network-scale deployments where a small percentage of inaccessible sites would otherwise create gaps in catchment coverage.
Managing hundreds or thousands of deployed monitoring devices requires dedicated tooling. Fleet management platforms provide centralised visibility of device health, battery status, connectivity, and data transmission across an entire deployment. Remote configuration changes, updating thresholds, transmission intervals, or alert routing, can be applied across the fleet without site visits.
Battery management is a key operational consideration at scale. Devices with 5–10 year battery life, tool-free replacement, and remote battery state reporting allow utilities to plan maintenance cycles efficiently and avoid unexpected data gaps from depleted devices.
When level data is clean and consistent, teams can:
- Detect surcharge earlier
- Understand wet weather response
- Identify chronic inflow and infiltration behaviour
- Prioritise investigations
- Justify capital and maintenance spend
The value here is not in the sensor, but rather in how confidently teams can trust and act on the data to make significant operational decisions. The shift of moving from reactive to proactive across the entire network.
Why sewer monitoring programs stall - and how to fix it
Sewer level monitoring has a higher failure rate than most other IoT deployments. Not because the technology is wrong, but because the physical environment is hostile and programs are frequently treated as generic telemetry rather than field-driven engineering challenges.
Why data quality starts at the pit
The decision to treat sewer monitoring as a software problem, solvable by better algorithms and more aggressive filtering, consistently produces the same outcome: an ever-growing complexity of analytics layered on top of fundamentally unreliable field data.
Filtering can mask the symptoms of a poor install. It cannot resolve them. A radar sensor positioned where its beam cone intersects a fixed ladder or pipe offset will produce inconsistent data at every level, in every weather condition, regardless of how sophisticated the downstream processing is.
The ceiling on data quality is set at the pit. Software can raise the floor, reducing noise, handling dropouts, flagging anomalies but it cannot raise the ceiling beyond what the physical installation permits.
Common reasons monitoring programs stall
Rushed installations – Mounting position not assessed for beam obstructions. Problems only become visible weeks later when data is reviewed.
Noisy untrustworthy data – Operators stop acting on alerts when the data is inconsistent. Once trust erodes it is very difficult to recover operationally.
Escalating maintenance – Sensors fouling, batteries depleting or frequent site visits eroding the cost case that justified the program in the first place.
Platform complexity – Systems that require specialist knowledge to configure or interpret create dependency on vendors and slow operational uptake.
Isolated monitoring – Single-site monitoring ticks a box that provides no comparitive insight. Programs fail to generate the network intelligence that justifies further investment.
No rainfall correlation – Level data interpreted without rainfall context leads to incorrect conclusions about the source and severity of network problems.
What good sewer monitoring looks like in practice
When the conditions for reliable monitoring are met, the program becomes a long-term operational asset rather than another platform that teams learn to ignore. The conditions are not complicated — they are consistent.
- The physical environment is assessed before sensor type and mounting position are decided
- Beam obstructions are identified and managed at installation, not post-processed
- Install time is kept short, avoiding confined space entry where possible, enabling scale
- Devices are deployed in clusters, not in isolation, to enable comparative analysis
- Level data is always correlated with rainfall records
- Alerts are configured conservatively at first to build operator trust before thresholds are refined
- Data quality is reviewed after the first significant rain event before the program is expanded
- Monitoring outcomes are tied to defined operational decisions — not generated as data for its own sake
Frequently asked questions
Related articles & guides
Got a monitoring challenge on your network?
Our engineers work with utilities across the globe. Tell us what you're dealing with and we'll tell you honestly what monitoring can and can't solve.
