To meet planetary targets, governments will need to better leverage data and AI

With just a few weeks to go until Glasgow opens its doors to world leaders at COP26 – the 26th UN climate change conference – hopes are high for impactful commitments to be made. Innovation is an integral part of delivering on these commitments, and governments will need to better leverage data and AI to effectively reach our planetary goals.

There is significant opportunity in using AI to meet our planetary targets

At the intersection of the environmental sector and the AI movement, we believe there is significant opportunity to meaningfully move both fields forward with help of one another. The Scottish Council for Development and Industry (SCDI) has recently called for a “Moonshot” to make Scotland a leader in Climate Tech, and the recently released UK National AI Strategy sets out a 10-year plan to make Britain a “global AI superpower”. As the fastest growing deep technology in the world and a critical component of the fourth industrial revolution, AI offers a step change when it comes to developing climate- and environmental solutions. As a certain level of digital maturity is needed to leverage AI, there is a clear need for governments and businesses who still work with paper-based processes and siloed data and infrastructure solutions to prioritise the digital transformation needed to jump on the AI train.

What types if use cases are we talking about?

AI is a broad category that can be applied to a vast range of problems, for example monitoring, identification, modelling, prediction and optimisation. To illustrate the benefit and potential of AI in the environmental sector, let’s take a look at some interesting use cases:

• AI driven monitoring through satellite imaging, can be used to model current states and future scenarios to help prioritise and guide restoration work. Deep learning or computer vision techniques can be used to improve classification of vegetation cover types which can improve rewilding efforts.
• Through the use of machine learning, energy generation and demand can be modelled in real time, improving smart grids and energy efficiency
• Automated data collection and decision-making at a local farm level provides precision agriculture where farms can optimise planting, spraying, harvesting crops, feeding, conception and detecting disease (both in livestock and crop). This means resource efficiency can be increased lowering the use of water, fertilisers and pesticides which otherwise find their way into rivers, oceans and insect populations.
• Smart sensors in beehives allow beekeepers to monitor their hive health from their smartphone, allowing for preventative action to reduce colony loss and strengthen bee populations. 
• Smart transport is in large enabled by AI, where machine learning algorithms are using car-sourced data for optimising routes, and autonomous vehicles use computer vision algorithms and deep neural net techniques.
• Invasive species is a serious challenge for upholding natural habitats as they alter the balance of ecosystems. For invasive animals and vegetation, computer vision and machine learning can be used for identification and monitoring impact in order to inform elimination.
• In Africa, wildlife is being protected by identifying poachers with the use of AI that analyses terabytes of video from drone aerial footage, in real time – day and night.
• Improving the case for renewables can be done in many ways, for example by driving down cost. Through the use of sensor technology and machine learning, remote analysis and predictive maintenance of solar and wind power generation can be enabled. Maintenance issues can be dealt with before they become critical.
• Air pollution is a big problem in many parts of the world, and air purifiers can be made a lot more efficient by introducing machine learning that analyses air quality data from the purifiers in addition to environment data and is able to adapt filtration efficiency in real-time.

The above use cases are only a few of many, and the positive impact AI can have on the environment is clear. In fact, a review of AI’s impact on the UN Sustainable Development Goals (SDGs) showed that up to 93% of the environmental targets could be positively impacted by AI. When talking about technological advancements, however, it is important to acknowledge the risks too, as the same study found AI could negatively impact up to 30% of the environmental SDG targets. This largely stems from the risks of high energy consumption and lagging AI legislation. Both are aspects we need to keep in mind when driving change.

Delivering innovative, digital solutions for public good

Delivering on the promises of AI to meet planetary targets will require designing desirable, feasible and viable propositions that leverage existing as well as new data sources. Storm ID has spent the past two decades helping public and private sector clients go through much needed digital transformations, making innovation possible. Lenus Health – a business developed by Storm, connects patient-generated data from e.g. wearables with pioneering AI, unlocking unprecedented insight, innovation and care transformation.

Lenus Health has rolled out across several health boards in Scotland, is clinically proven to improve patient health outcomes, and is now in the midst of national and international expansion. Fundamental to its success, is the interoperability of the platform, the user-centric design and the enabling of preventative care. To drive public good, Lenus Health is built with open APIs to invite the healthcare sector and 3rd party developers to leverage the platform for further innovation.

This way of working – pushing the boundaries of technology and making it accessible for collaboration – is what we need in the environmental sector too. In order to hear more about Lenus Health’s learnings applicable to the climate sector, sign up to the free one day conference Scotland’s contribution to COP26 held on the 26th of October where co-founder Paul McGinness joins in to share his views.