With nature-based stormwater systems gaining greater acceptance, the need for long-term maintenance grows. Jac van Tuijn hears from Frédéric Cherqui about the potential role of low-cost sensors.

Traditional underground drainage infrastructure has offered the benefit of letting excess rainwater flow out of urban centres quickly. Alongside this, more and more cities have begun to introduce stormwater control measures (SCMs) to catch and infiltrate excess stormwater and help delay peaks that overwhelm networks. Popular measures are ponds, sunken grass swales, infiltration trenches, green roofs or porous pavements.

Such systems are often nature-based solutions, above ground, highly visible and accessible to the public. This means that incorporation in the landscape and the potential public use of an infiltration or retention area represent important aspects of such measures. Moreover, there has always been a focus in research on the design, with less of a focus on long-term performance and maintenance.

According to Frederic Cherqui, Associate Professor at University Lyon 1, France, it is time to study these aspects more closely and get a better understanding of the medium- and long-term performance of SCMs. “Asset management of SCMs is now a concern for early adopters and may become a strong barrier to the adoption of such systems,” he says. The focus of his research includes improvement of the monitoring of nature-based stormwater systems, and he coordinates the EU-funded Mind4Stormwater project, which focuses on developing innovative monitoring solutions to maintain the overall performance of these systems in the long term.

“Suitable monitoring is required at the right place and right time, to intervene before malfunctions occur”

Proactive maintenance

Nature-based stormwater systems are praised for their ability to mitigate run off and keep the rainwater inside the city. “It is an attractive proposition that has powered the rise of nature-based solutions in general,” Cherqui states. “At the same time, the promise of low maintenance requirements for such assets still presents a problem.” He sees that management is all too often based on a run-to-failure approach. He urges proactive, long-term-orientated maintenance. “Active asset management is needed to control stormwater. And there is an opportunity to learn from the 50-plus years of experience of sewer asset management,” he adds.

Low-cost sensors

Cherqui notes that a major limitation is the ability to monitor the design intent of SCMs. Suitable monitoring is required at the right place and right time, to intervene before malfunctions occur. “Doing so has traditionally been very expensive and labour-intensive,” he continues.

As part of the Mind4Stormwater project, Cherqui has been involved in a study on the use of low-cost sensors for SCM control. The project started in 2018 and has looked at how to optimise performance relative to the operating conditions and to the state of the maintenance of the system.

Cherqui has noted with co-authors in publications on the project that low-cost technology often needs parts from different providers, and then requires customised programming. The modularity and open-source nature of such tools offer major advantages, but this comes with a cost in terms of required skills and build time. “Low-cost or frugal monitoring systems shouldn’t be considered as affordable replacements of traditional monitoring systems: they offer a total control of the whole monitoring chain and will question our monitoring practices,” he says. There can be questions around the reliability of a low-cost sensor approach but, inspired by supervisory control and data acquisition (SCADA) systems, can open the way for a new range of possibilities for processing and analysing real-time, onsite data. The opportunities increase if data is shared with other actors such as departments involved in drainage, landscapes or street sweeping.

“Our challenge is to make the most of the ongoing Internet of Things innovation”

Sensor testing

In order to address reliability, the Mind4Stormwater project has included the development of a platform and testing protocol to assess the suitability of low-cost sensors. According to Cherqui, this is not only important for nature-based SCMs, but for the water sector as a whole. “Testing sensors is not only mandatory before their use, but it also enables them to assess their full potential: measurement time, accuracy, precision, range, energy consumption, etc.,” he says. Sensors need to function well across a range of environmental conditions. Some sensors include software or hardware systems to compensate for the change of environmental parameters, such as pressure sensors that integrate pressure compensation based on water temperature.

One major area of application is in the monitoring of water levels. “We built a platform to test the performance of water level sensors that use different measurement principles,” Cherqui explains. “We assessed the performance of three widely used low-cost sensors: laser-ranging, ultrasonic-ranging, and pressure. Our main results showed that the ultrasonic sensor offered the best price-to-accuracy ratio and the pressure sensor provided the highest accuracy while still at a very low cost. The research led to a testing protocol that is fully customisable to the sensors being tested and all the following parameters,” says Cherqui.

IoT innovations

The ultimate goal of a reliable and functioning monitoring system is to support urban water management. For nature-based measures, this is not limited to the hydraulic functioning, as they often serve more functions. This makes it important, for example, to understand the response to a storm event or enable early detection of abnormal behaviour related to component failure. Through the Mind4Stormwater project, Cherqui is currently involved in developing an innovative expert system to guide utility managers in selecting the best operation and maintenance actions for each SCM.

With the Mind4Stormwater project due to conclude later this year, Cherqui also mentions the development of a monitoring strategy for Australian case studies. “This will support development of the expert system, which will then be adapted for French cities, and a version then drafted aimed at European cities,” he says. “Our challenge is to make the most of the ongoing Internet of Things [IoT] innovation, and especially to develop and use existing low-cost sensors to enable more monitoring.”

Towards real-time control

Innovative monitoring systems not only provide real-time access to data, but also present a game-changing strategy. The possibility to know in real time the state of a system and to trigger manual or automatic actions based on its state and other parameters such as rain prediction is becoming a reality even for small systems such as SCMs. “I can see a near future where assets are continuously optimised to provide the best performance at the lowest cost,” concludes Cherqui.