In southwest England, artificial intelligence will be used to proactively predict water pollution.
A pilot project in Devon is expected to help prevent water pollution in the seaside resort of Com Martin and make it a suitable place for swimming.
As part of the project, about 50 sensors will be installed in rivers and fields, and AI will combine that data with satellite imagery of local land use to predict when local river systems will be most vulnerable to pollution. To do.
Farmers would then be asked to refrain from applying fertilizer when the river is most polluted.
The project, led by computer systems company CGI and map specialist Ordnance Survey, is being piloted in the North Devon Biosphere Reserve, a protected area that includes natural habitats, farmlands and small towns.
According to CGI, the accuracy of the AI technique was more than 90% during test runs. The company’s chief sustainability officer, Matty Yeta, said the data generated by the sensors feeds into AI technology, which “can tell us where these incidents are happening and when they’re actually happening. “We will be able to develop a prediction mechanism,” he said.
Ordnance Survey’s Donna Lindsay said: “We can start training models with the data so that we can understand when a pollution event occurred and whether it is related to a particular region.” Were there any specific rainfall events?”
This happened amid concerns about the water quality of Com-Martin’s baths. In 2018 and 2019, the resort received a “poor” rating and people were advised not to swim.
“There’s a lot of anxiety in the community about what will happen if the bathing water permit is revoked,” said Andy Bell of the North Devon Biosphere Reserve.
“It will affect cafes, restaurants, B&Bs…People want to come to clean places to have fun.”
Yeta said he hopes the project will scale and be rolled out to different parts of the UK.
