Artificial intelligence is set to play a crucial role in predicting and preventing water pollution.
A pilot project in Devon, south-west England, involves the deployment of sensors in rivers and fields to monitor local river conditions, rainfall, and soil in the hope of improving water quality in the seaside resort of Combe Martin.
The data collected by these sensors will be combined with satellite imagery of the area's land use using AI technology.
This integration will then enable the prediction of vulnerable periods for the local river system, such as during agricultural runoff, facilitating proactive measures like delaying fertiliser application on farms.
Mattie Yeta, chief sustainability officer at computer systems company CGI, which is heading the project, told the BBC: “We’ll give (the AI) the history... we’ll give it all of the geographic information, as well as data sets from the sensors for it to learn and develop the predictive mechanisms to be able to inform where these incidents are occurring and indeed when they will take place.”
CGI is collaborating with mapping experts Ordnance Survey on the scheme,
with the pilot project being conducted within the North Devon Biosphere Reserve – a protected area covering 55 square miles that encompasses significant natural habitats, farmland, and small towns.
The project’s main objective is to enhance the water quality of Combe Martin, a seaside resort that has long struggled with bathing water concerns.
Andy Bell from the North Devon Biosphere Reserve, said: “It would impact on the cafes, the restaurants, the bed and breakfasts... people want to come to a clean place to enjoy themselves.”
The River Umber, responsible for the majority of pollution, flows into the sea beside the beach through a corridor of algae growth.
Efforts to improve water quality start with the collection of real-time information, facilitated by the installation of a floating water sensor in the river. The sensor continuously transmits data on key indicators of water health.
Glyn Cotton, CEO of Watr, said about its capabilities: “If sewage was being discharged upstream we would see spikes in things like ammonia and pH, and we can then cross-reference that with temperature and dissolved oxygen levels.”
Donna Lyndsay from Ordnance Survey added: “We can start training the model using data to get it understanding... was there, for example, a particular rainfall event that washed it all off?”
The ultimate aim is for the AI to provide actionable recommendations based on the data analysis, including that if dry soil and heavy rain are predicted farmers could be advised to refrain from applying more fertiliser to prevent runoff.
