SSE Renewables teams up with Microsoft for AI-led puffin conservation project on Isle of May

Green power firm SSE Renewables has teamed up with Microsoft on an artificial intelligence (AI) project trial in Scotland that could speed up the time it takes to bring new sources of renewable energy onto the grid.

Scotland is renowned for having some of the most ambitious renewable energy usage targets in the world, having pledged in 2011 to have 100% of its electricity demand met by green power sources by 2020.

Figures published by the Scottish government in March 2021 revealed that the country missed its 2020 target by a few percentage points, with the country managing to meet 97.4% of its electricity demand using renewables.

Work on decarbonising its economy and ramping up its use of renewables continues apace regardless, but one of the biggest barriers to bringing new sources of renewable energy online is ensuring that the build-out of the required power generation infrastructure does not impact the environment or native wildlife populations.

This is a challenge SSE Renewables found itself facing, and has sought to address with the help of Microsoft, cloud consultancy Avanade, and government conservation agency NatureScot.

The group have recently completed a trial technology project that relies on AI and machine learning tools, combined with specialised cameras and image recognition technology, to identify and count puffin colonies on the Isle of May in Scotland.

SSE Renewables operates an 84-turbine offshore windfarm in the north of Scotland, 13km off the coast of Caithness, which is responsible for generating 588MW of energy a year, enough to power 450,000 homes.

Part of this project was geared towards potentially identifying ways in which it could improve its environmental monitoring procedures to ensure its energy-generating activities in this area are not affecting the local puffin population.

“We wanted to improve our environmental monitoring so we can be more proactive in that area,” said Oliver Abell, account manager at SSE Renewables’ Engineering Centre. “It’s part of a digital transformation at SSE Renewables that Microsoft and Avanade have been helping us with.

“They [Microsoft and Avanade] showed us a lot of technology that could help us to meet our objectives and worked with our staff to show them how they could utilise that tech.

“There is no way SSE Renewables could have started the AI puffin counting project on our own – we are an energy company, not a technology company. Microsoft and Avanade have been critical pieces of this project.”

About 80,000 puffins were recorded as living on the 140-acre Isle of May in March 2020, making it the third largest colony in the UK.

Puffins are a species of concern for conservationists because of the relatively low numbers of them that exist in the wild, and the fact that female birds lay only one egg a year.

To keep tabs on population numbers, specially-trained rangers are usually deployed to physically scout about the puffins’ burrows to count the birds and the number of eggs they have laid.

The aim of the project is to make this work less labour-intensive and physically demanding for the rangers, while minimising the risk of habitat disruption for the puffins by limiting the amount of human interaction they encounter.

“This particular area on the Isle of May attracts many puffins each year to breed and it’s key to ensure that with the planning of any new windfarms, this is not interrupted,” said Simon Turner, CTO of data and AI at Avanade.

“To monitor the puffin population on the island, NatureScot would normally send people with clipboards to sit for hours, marking down how many puffins they saw. With SSE Renewables, we saw an opportunity to use technology to make this process more accurate, more efficient and less invasive for the puffins.

“Using cameras and AI, we are now able to count the number of puffins and monitor their burrows all day, every day, without going near them.”

The project saw SSE Renewables and Avanade set up four cameras in stainless-steel boxes on the island that provide a live feed of the puffins’ activities, between late March and mid-August. This period is when the birds typically return to the island for breeding purposes after eight months away at sea, before leaving again before autumn arrives.

The data captured by the cameras is stored in a Microsoft Azure public cloud data lake, and relies on the serverless Azure Kubernetes Services to help manage the information it collects.

Avanade and SSE Renewables have drawn on this setup to incrementally train an AI using an image recognition model on increasingly complex objects, including pictures of puffins, so that it understands how to spot a bird amidst the surrounding landscape.

“When the cameras on the Isle of May are turned on, the AI will be able to spot the puffins, separate them from background images such as rocks, and track them, frame by frame, as they move around,” said Microsoft in a blog post.

Expanding on this, Turner said the AI is capable of drawing a box around each puffin it spots and giving them a unique identification tag so it knows not to double-count the birds.

“When the camera moves to the next frame, the AI understands that the puffin closest to a particular box is the same puffin, it’s just taken a step to the left or right, so it redraws the box around the bird,” said Turner.

“This happens over and over again for every frame of the footage. Even if the puffin flies out of the frame, the AI system will recognise where it flew out of sight, and attempt to track it again if it comes back. That’s how we can track and count individual puffins.” 

SSE Renewables’ Abell said it can take up to 10 years for green energy firms to secure consent for new windfarms, and the hope is that projects like this will speed up the time it takes to bring new sources of renewable energy generation online.

This, in turn, will help Scotland and the rest of the UK achieve its green energy usage goals more quickly, and plans are already under way to look at other ways to deploy this technology in support of that.

“We could monitor salmon to make sure they are able to migrate in rivers, for example,” Abell added. “This technology could work in any environment in which you want to be able to monitor a species and be hands-off, either because it’s too remote or because you don’t want humans interfering in that environment.”