River level monitoring: how IoT sensors deliver real-time flood warning for UK sites

TL;DR: River level monitoring uses IoT sensors to measure water height in rivers, channels, and watercourses in real time, triggering alerts before floodwater reaches buildings, roads, or critical infrastructure. With 6.3 million properties at flood risk in England and climate change pushing that figure toward 1 in 4 by 2050, connected monitoring is the fastest route from reactive sandbagging to proactive early warning.

Key takeaways

• 6.3 million English properties sit in flood-risk areas, and surface water risk has risen 43% since 2018 (Environment Agency, 2025).
• 60 GHz mmWave radar sensors like the AQUAIOT Radar measure river levels to within +/-2 mm without touching the water.
• Ofwat has approved a record £104 billion AMP8 investment for 2025 to 2030, with £12 billion ringfenced for storm overflow reduction.
• A monitoring system that pairs radar sensing with LoRaWAN or cellular telemetry delivers alerts in minutes, not the hours a manual gauge check requires.
• The Flood and Water Management Act 2010 places Lead Local Flood Authorities under a duty to assess and manage local flood risk, making continuous monitoring an operational advantage.

Contents

• What is river level monitoring and why does it matter in 2026?
• How many UK properties face river flood risk?
• How do IoT sensors measure river levels in real time?
• Which sensor types work best for river level monitoring?
• How does river level monitoring connect to the UK flood warning network?
• How should you choose a river level monitoring system?
• The full AQUAIOT flood monitoring capability stack
• Frequently asked questions

What is river level monitoring and why does it matter in 2026?

River level monitoring is the continuous measurement of water height in rivers, streams, channels, and open watercourses using fixed sensors that report readings to a cloud dashboard in real time. It matters because even a 200 mm rise in a watercourse near a building, road, or drainage outfall can trigger flooding that costs tens of thousands of pounds to remediate. In 2026, the combination of intensifying rainfall patterns, ageing drainage infrastructure, and tighter regulatory expectations makes river level monitoring a core operational requirement for councils, utilities, estates teams, and housing providers across the UK.

The Environment Agency already operates a public network of monitoring stations across England, feeding data into the Check for Flooding service. But that network covers main river channels. Smaller watercourses, culverts, and site-specific flood pathways often sit outside it. That is where local IoT monitoring fills the gap, giving asset owners their own alerting layer rather than relying on national warnings alone.

Connected monitoring replaces the manual gauge board with a sensor that reads water height every few minutes, sends data over cellular or LoRaWAN telemetry, and fires an alert when a threshold is breached. The shift from periodic inspection to continuous visibility is the difference between reacting to flood damage and preventing it.

How many UK properties face river flood risk?

The Environment Agency’s national assessment for 2024 to 2025 puts the figure at 6.3 million homes and businesses in areas at risk from flooding in England, equivalent to 1 in 5 properties.

Of those, 2.4 million are at risk from rivers and the sea, while 4.6 million face surface water flood risk, a figure that has risen 43% since the previous assessment in 2018.

Climate projections push the numbers higher. By mid-century the proportion at risk climbs to 1 in 4 properties, with river and sea risk reaching 3.1 million and surface water risk hitting 6.1 million. The financial toll is already visible: the Association of British Insurers reported that domestic flood claims reached £312 million in 2025, a 38% year-on-year increase, with the average homeowner payout rising 60% to £30,000.

For estates managers and council flood officers, those numbers translate into a direct case for continuous watercourse monitoring: if you can see the rise in a watercourse 30 minutes before it tops a bank, you can close valves, move assets, and alert residents rather than filing an insurance claim afterwards.

How do IoT sensors measure river levels in real time?

An IoT river level monitoring sensor sits above the water surface, either on a bridge parapet, a bracket over a channel, or inside a stilling well, and fires a signal downward to measure the distance to the water. The sensor calculates the current water level from that distance reading and transmits it to a cloud platform at set intervals, typically every 5 to 15 minutes. When the level crosses a pre-set threshold the platform fires an alert by SMS, email, or voice call to the designated responders.

The core components of a river level monitoring installation are the sensor head (radar, ultrasonic, or pressure), a communications module (cellular 4G/LTE or LoRaWAN), a power source (solar panel, battery, or mains), and a cloud dashboard that stores readings, draws trend graphs, and manages alerts. LoRaWAN is popular for remote riverbank locations because it offers long range (up to 15 km line-of-sight) with very low power draw, giving battery-only deployments a lifespan measured in years rather than months.

The AQUAIOT Radar uses 60 GHz mmWave technology for non-contact river level monitoring, measuring to within +/-2 mm accuracy. Because it operates at millimetre-wave frequency, it reads through foam, vapour, condensation, and surface turbulence that would confuse lower-frequency sensors, making it reliable in the wet, enclosed, and debris-prone conditions typical of UK rivers and culverts.

Which sensor types work best for river level monitoring?

Three sensor technologies dominate river level monitoring in the UK: radar, ultrasonic, and submersible pressure transducers. Each has trade-offs in accuracy, maintenance burden, and deployment cost that determine which fits a given site.

Radar (mmWave). Non-contact, mounts above the waterline, unaffected by temperature swings or debris in the flow. The 60 GHz band used by the AQUAIOT Radar delivers stable, accurate readings over long distances and penetrates condensation and vapour. Best for exposed river channels, culvert inlets, and environments where the sensor must survive without routine maintenance. ATEX-rated variants suit hazardous atmospheres near treatment works or chemical storage.

Ultrasonic. Also non-contact, measuring the time-of-flight of a sound pulse reflected off the water surface. Cheaper than radar at short range, but accuracy degrades in high wind, heavy rain, or where turbulence breaks the return echo. Typical accuracy is +/-10 mm at ranges under 10 m. Suited to calm, sheltered channels with clean sightlines.

Submersible pressure. A transducer sits at the bottom of the channel and measures the hydrostatic pressure of the water column above it. Reliable in deep or enclosed watercourses but requires physical contact with the water, which means silt build-up, biofouling, and periodic retrieval for cleaning. Typically used as a secondary reference or where overhead mounting is impossible.

For most UK council and estates deployments, radar river level monitoring offers the best balance of accuracy, reliability, and total cost of ownership. The absence of moving parts and water contact means less maintenance and longer service life in field conditions.

How does river level monitoring connect to the UK flood warning network?

The Environment Agency, Natural Resources Wales, and SEPA operate the national flood warning system, issuing flood alerts and flood warnings based on data from their own gauging stations. The Flood Forecasting Centre, a partnership between the Environment Agency and the Met Office, publishes five-day flood forecasts for England and Wales. These national warnings cover main rivers and known flood plains, but they do not extend to every ordinary watercourse, culvert, or site-specific drainage pathway.

Local watercourse monitoring fills that gap. Under the Flood and Water Management Act 2010, Lead Local Flood Authorities (typically upper-tier councils) have a statutory duty to investigate and manage local flood risk. A council that deploys its own river level monitoring network on ordinary watercourses gains advance warning that the national system does not provide, plus a timestamped evidence trail for incident reporting and insurance responses.

The FCERM Strategy for England sets the long-term direction: a nation ready for, and resilient to, flooding by 2100. That ambition depends on denser monitoring coverage than the current national network alone can deliver. IoT level sensors, deployed and managed locally, are how councils, water companies, and estates teams extend that coverage to the watercourses closest to their assets.

How should you choose a river level monitoring system?

Selecting the right river level monitoring system means matching the sensor technology, connectivity, and alerting capability to the site conditions and the operational outcome you need. These five criteria separate a system that delivers actionable data from one that generates noise.

1. Accuracy and range. Confirm the sensor resolves the water level change that matters to your site. For a river channel where a 50 mm rise triggers bank overtopping, a sensor accurate to +/-10 mm is the minimum. The AQUAIOT Radar’s +/-2 mm resolution gives a wide margin on that requirement.

2. Environmental resilience. Rivers throw condensation, debris, turbulence, and freezing spray at sensors. Radar performs reliably across all of these. Ultrasonic sensors struggle in high wind or heavy rain. Pressure transducers foul in silty flows. Choose the technology that survives your worst conditions, not your average day.

3. Connectivity. Remote riverbank sites may lack mains power and mobile signal. LoRaWAN gives long range on low power. Cellular (4G) gives higher bandwidth for frequent readings. Some sites need both. Check coverage before fixing the sensor type.

4. Alerting and integration. The system must fire threshold alerts by SMS, email, or voice to the right people within minutes. It should also integrate with existing estate management or SCADA systems via API so that river level monitoring data feeds the same dashboard your team already watches.

5. Deployment and maintenance. Retrofit-friendly mounting (bridge parapet, bracket, or pole) avoids civil works. Non-contact sensors like radar avoid in-water maintenance. Battery or solar power avoids mains cabling. The lower the ongoing maintenance burden, the more sites you can cover within budget.

The full AQUAIOT flood monitoring capability stack

Flood risk does not stop at the riverbank. Water moves from catchment to channel to drain to building. AQUAIOT covers each layer of that chain with a specific product or service, all feeding the same cloud dashboard.

Prevent: Smart Water Butt

The AQUAIOT Smart Water Butt brings IoT to rainwater harvesting. Rainfall-aware logic pre-empties storage capacity before storm events, reducing runoff into drains and watercourses. Relevant wherever SuDS attenuation or stormwater management is part of the flood resilience plan.

Outdoor level sensing: AQUAIOT Radar

The AQUAIOT Radar is the primary river level monitoring sensor. 60 GHz mmWave, non-contact, +/-2 mm accuracy. Deploys on rivers, tanks, and open channels. ATEX variants cover hazardous environments near treatment or chemical sites.

Sewer and drain: Radar for underground assets

The same Radar sensor applied in manholes, wet wells, and CSO chambers delivers the sewer monitoring service. It tracks levels in enclosed underground spaces where condensation and gas would blind lesser sensors. Continuous sewer level data feeds directly into overflow prevention and pollution incident response.

Indoor leak: NOAH Multifunction Leak Sensor

When floodwater enters a building, or when internal pipework fails under pressure, the NOAH Multifunction Leak Sensor catches the first drop. A NOAH Leak Sensing Membrane triggers on contact with water, firing an alert via the KAIROS Portal within minutes. LoRaWAN Class A radio, up to 10-year battery life, and a small form factor make it retrofit-friendly for apartments, plant rooms, and commercial buildings.

Flow anomaly: Clamp-on Ultrasonic Flow Meter

The Clamp-on Ultrasonic Flow Meter monitors pipework without cutting into it. DN8 to DN100 pipe range, non-invasive, retrofit-friendly. Night-flow profiling and anomaly detection catch leaks that compound flood damage by wasting treated water while the estate recovers.

Cloud, alerts, integration

Every sensor in the stack reports to the AQUAIOT Cloud and the KAIROS Portal: role-based dashboards, threshold alerts (SMS, email, voice), trend graphs, exportable records, and API integration with SCADA, BMS, and CAFM systems. One login, every site, every asset class, from river level monitoring stations to indoor leak sensors.

Frequently asked questions

What accuracy do you need for river level monitoring?

For most UK flood-warning and watercourse-management applications, a sensor accurate to +/-10 mm or better is sufficient to detect the level changes that trigger bank overtopping or drain surcharge. The AQUAIOT Radar delivers +/-2 mm, giving a wide safety margin and the resolution to spot gradual trends as well as sudden rises. Higher accuracy also improves the quality of trend data used for long-term flood risk assessment.

Can river level monitoring sensors work without mains power?

Yes. Most IoT level monitoring deployments run on battery or solar power combined with LoRaWAN telemetry, which draws very little energy per transmission. A solar-powered radar sensor on a riverbank can operate for years without a site visit, making it practical for remote watercourses where running mains cabling would be prohibitively expensive.

How does river level monitoring differ from flood warning?

River level monitoring is the continuous measurement of water height at a specific point. Flood warning is the operational decision, based on monitoring data, forecast rainfall, and catchment models, to issue an alert that flooding is expected. Monitoring provides the data; warning provides the action. A local monitoring network gives an organisation its own data source rather than relying solely on national Environment Agency warnings, which may not cover every watercourse near its assets.

By Gianbattista Porru, Digital and IoT lead at AQUAIOT

Last updated: 3 June 2026