PDF Basket
Our water and sewage systems are under constant strain. Increasingly severe and unpredictable storms can overwhelm infrastructure, resulting in raw sewage flowing into rivers and lakes. Improved monitoring systems allow prompt action to be taken where needed, ensuring that water quality is maintained, and citizens protected.
The EU-funded DWC project demonstrated how leveraging smart digital technologies could enable more efficient and timely water monitoring. Prior to the project, the adoption of such technologies had been limited by a lack of viable business cases and concrete evidence of their effectiveness.
Demonstrating digitalisation’s added value
The project brought together five major European cities – Berlin, Copenhagen, Milan, Paris and Sofia – as test cases for new monitoring solutions. “These cities all have different challenges to address,” explains project coordinator Nicolas Caradot from Kompetenzzentrum Wasser Berlin in Germany. “The idea was that we would identify and develop solutions that address these specific problems.”
One key challenge in city centres with older infrastructure is that sewage and stormwater systems are often combined. This can mean that sewage might be released into rivers during storms – as seen during the Olympic Games in Paris, when swimming events in the Seine had to be rescheduled due to water contamination.
For the cities that took part in the project, key issues included a need for better management of sewage network flows during storms, monitoring of river water quality, optimisation of operational costs and maintenance investments, and safe treatment and reuse of municipal wastewater for agricultural irrigation.
“These were the identified key challenges,” says Caradot. “We then gathered 15 innovators and worked on developing digital solutions. These solutions were tested in at least one of our cities. Our aim was to show in practice the added value of digitalisation.”
Automated systems successfully trialled
Successfully trialled solutions included a sensor system that informs operators almost immediately if there is a risk of water contamination. The sensor system was tested in the Seine in Paris, as well as in Berlin and Milan.
“Before this, you would have to take a sample to the lab and wait 24 hours for a result,” adds Caradot. “With this technology, you have a result within a few hours, directly on-site.”
A key limitation of standard laboratory methods is their inability to accurately measure bacterial concentrations in urban water supplies. This makes them unsuitable for performing microbiological risk assessments in areas impacted by urban wastewater pollution.
The new sensor addresses this issue by providing both planktonic (free-floating) and comprehensive faecal indicator bacteria counts (including those aggregated on faecal particles), leading to higher-accuracy risk assessments and better public health protection.
To accompany this, the project team also developed a machine learning-based early warning system for improved bathing water management. The system was able to predict water quality a few days ahead of time with an accuracy level of 95 %, and provide advance warnings of water quality degradation. “We gathered data on features like rainfall, sewage flows and treatment quality, and applied a machine learning algorithm to predict water quality, limiting the need for physical measurements,” explains Caradot.
On the sewer network side, a low-cost monitoring solution was developed to enable utility operators to monitor a vast number of combined sewer overflow outlets. Sensors were also developed to address problematic connections between sewage systems and stormwater systems.
Helping to shape future policy
Since its completion, tools developed by the project have already been put to use in Europe. These include an early warning system for safe water reuse in agriculture, which is currently being used in Italy. A decision support system that uses algorithms for better stormwater management is now in operation in Copenhagen.
“There were also a number of solutions pioneered in this project that were not ready to be brought to market, but which nonetheless showed great potential,” notes Caradot. “These included the enzymatic sensors for water quality monitoring, and the machine learning-based early warning system for bathing water quality.”
Caradot and his team are currently applying for further EU funding. The aim is to integrate these two innovations – water quality sensors and computer-based modelling – to bring a combined tool to market. “We are also looking to apply this in other fields as well, such as aquaculture, agriculture and drinking water protection,” he adds.
The DWC project also worked closely with other EU-funded initiatives, including SCOREwater and Fiware4Water, to secure a long-lasting impact on water quality monitoring in Europe. This also resulted in the publication of a policy brief which identifies legislative gaps and offers key recommendations. “This is a reference document not just for us, but for other researchers and policymakers,” says Caradot. “The aim is to help shape future policy.”