How Can AI And Geodata Be Used To Support Biodiversity?
We are dependent on biodiversity for many things – food, water, medicine, a stable climate and economic growth among others – but are not very good at looking after it.
Supporting and sustaining biodiversity is a key part of global sustainability, however there is not a one-size-fits-all approach.
Drew Purves, Sustainability and Biodiversity Co-Lead at Google DeepMind, Google’s AI arm, and Robert Hoddenbach, Global Director of Climate & Nature Markets at geodata leader Fugro, both believe that technology can have a positive impact on biodiversity.
Can AI be a tool for sustainable good?
Before joining Google Deepmind in 2015 and after gaining his PhD from the University of York, Drew spent time at Princeton University, the University of Cambridge, Microsoft Research, the British Ecological Society and University College London. Having begun his studies in ecology and biodiversity, his continued focus has always been on how technology can positively impact the environment.
“In the long run, I believe that AI is one of the most promising technologies for addressing environmental challenges,” says Drew.
After taking some time to work in the fundamental AI research teams at Google Deepmind, Drew is using those foundations to benefit the environment.
“My time in the engineering support group within Google DeepMind gave me a bit of a feeling for the strengths and weaknesses of different methods and what it took to apply AI and develop it,” he explains.
“What I'm trying to do in my current role is coming back into the world of environmental sustainability, thinking about how we can apply those lessons. I'm coming back to the same topics that I’ve worked with before like biodiversity, conservation, nature-based solutions to climate issues, and thinking how we can approach that using AI.”
The team at Google Deepmind are world leaders in AI development, and Drew’s mission is to ensure that it works for sustainable good.
“We're trying to get as many of the smart researchers and engineers as possible pointing their talents towards the environment and other sustainability challenges, which are the most important challenges of our time.”
AI use case: Weather modelling
One example of how technology can positively impact sustainability is Google Deepmind’s project using AI in weather modelling.
“We now have the world's best weather predictions coming out of one of these deep neural nets that we use,” he says. “By best, we mean if you pick a certain time window into the future, it's the most accurate prediction or you can say for a given level of accuracy, we can predict further out into the future.”
This feeds into many different areas of sustainability, including renewable energy generation, protecting energy grids, sustainable agriculture and many more.
As well as the applications presenting lots of avenues for sustainable action, the predictions themselves are a lot more sustainable than you might think.
“Whereas all the previous models needed supercomputers, we can make those predictions on a laptop,” Drew celebrates. “When you get AI right, you can not only push the state of the art, but you can do it on much less compute.”
This makes it far more accessible and allows downstream users to use the technology.
AI use case: Boosting electrical grid efficiency
Drew’s team is also looking at algorithms to balance electricity grids.
“It’s very explicit energy work,” Drew says. “If we can use AI and deep learning to solve those algorithms much more quickly, that opens up all sorts of possibilities for making electricity grids more efficient – especially as we add in these variable sources of things like solar and wind.”
The vision is that AI could reduce energy use and carbon emissions from the electricity grids.
AI use case: Acoustic monitoring
Acoustic data is essential for the understanding of biodiversity, and Google Deepmind is using AI to monitor it.
“We’ve developed small networked devices that you can deploy for monitoring,” Drew says.
“They’re little microphones that you can put out in the field that create an enormous amount of acoustic data.”
Drew’s team has created an adaptable foundational model over the data that allows conservationists to use it for many purposes, including identifying species.
“Say people that are working in the tropics have very long lists of very rare species but not many recordings. We’ve developed automatic ID systems by adapting that foundational model with a small amount of data. So that's a key example of using AI to bring in more observations and data from nature, or monitoring nature.”
Google Deepmind is also developing a similar project with motivation activated cameras. The foundational models support the identification of the animals turning up in those images and vastly reduce the amount of data that needs to be manually trawled through.
Drew narrows AI’s use in acoustic monitoring cases into three stages:
- Deploying AI to bring in more data and monitor nature
- Integrating all that data together to be able to derive actionable information out of it
- Combining the information with other geographic information to predict where species are
“One of the projects that we're interested in is species distribution modelling,” says Drew, discussing the second category.
“That's actually quite a bottleneck at the moment. If you pick a point on the map, we often don't know which species are there. If we don't know what's there, we don't know what to protect, we don't know what to monitor.”
He highlights the importance of the third category – using that information.
“There is a danger when it comes to actual implementation and decision-making because of information overload. At that point you've got predictions where millions of species are at fine grain every day – what can you do with this?
“The active deployment of AI at that third stage to support decision making, assists groups of individuals in reaching the right decisions and developing the right policy.”
Fugro: Pioneering geo data for the climate and nature
Fugro is a company that has always supplied all kinds of geographic information – Robert says that the company “describes the earth.”
Like Google Deepmind, Fugro is turning its attention to use cases in climate and nature.
“My role involves looking for the business of applying geo data in the space of climate in nature,” Robert explains.
“That includes the use of geo data to understand climate change, adapt to climate change, do adaptation programmes or projects and the use of geo data to understand the status and the preservation and restoration of nature, but also the value of nature.”
In practice, what does this look like? In its own operations, Fugro’s sustainability efforts include using cleaner energy to power its ships, building on crewed surface vehicles to reduce the size of the ships and its footprint whilst working, as well as smart building management, digitisation and electrification.
“We go beyond corporate social responsibility to creating shared value,” Robert says. “For us, that means that we are focusing on our strengths that have built up over 60 years to do the right thing with it, to make sure that we create value not only for ourselves and for our shareholders and our employees, but also for society.
“I'm looking for projects using geo data for climate and nature related topics that ensure that our stakeholders get value out of what we do, civil society gets value out of what we do and we make a profit out of it. As a result, we have an economic mechanism that will perpetuate.”
Gathering data is crucial to be able to support biodiversity – understanding is the first step to positive impact, and data is crucial to understanding. Fugro is the expert in how to measure often complicated things, and at scale. This includes coastlines and areas the size of whole countries.
“Measuring areas that are complex or at scale, that's where we differentiate ourselves because we have a long standing history in the space and we have that big toolbox for technology,” Robert explains.
“Geo data is crucial in understanding climate change. Whilst there is a fair understanding of the large models and mechanisms of climate change, on a local scale there are many mechanisms we only partially understand or only understand in our abstract level. If you want to mitigate or adapt to the situation, you need to have a very, very detailed understanding in the local area. This is where we come in.”
Fugro’s risk modelling and mapping of things including coastal erosion, rising sea levels and the acidification of the ocean informs governments, companies and organisations.
Geo data use case: Ocean mapping
The biodiversity of the seas is often underestimated – half of the planet is covered in water, and more than 90% of the Earth’s biodiversity is in the oceans. The vast majority of heat is absorbed by the oceans, as well as them being a key capturer of carbon. Fugro’s expertise lies in water related mapping, particularly flood management and oceans.
“Our inland water related projects include flood management and ground control. There's always too much water or too little water on land – that's a problem, so we're focused on that,” says Robert.
On the other end of the spectrum is Fugro’s deep sea work. A large part of the ocean is unmapped, so Fugro is working to explore and chart it.
“And then in the middle is the coastal zone, which is technically the most complex area. There are big challenges in that space – it changes so fast because of the water and the sea level rising, but it also has the highest impact because 40% of the population is living on the coastline.”
In practical terms, this informs deep sea cable positioning, which is crucially important for the development of AI.
“A lot more cables are being laid, so you need to know where to lay it. If there's an underwater mountain in the way – something that is thousands of metres high on the seafloor – you have an issue.”
What is the future for technology’s support of biodiversity?
Technology in sustainability is ever-developing as governments, companies, organisations and individuals work to become more sustainable, however necessary.
“People are realising we can't reduce environmental sustainability just to climate and carbon,” Drew says.
“Leading institutions including COP28, the World Economic Forum, World Resources Institute and many more are pointing out that we need to think of climate, nature and people together. The awareness of the importance of nature as part of this integrated approach that we need to take to environmental sustainability is increasing quite rapidly.
“So then it's kind of natural for us to start thinking about what we can do with AI to help nature protection explicitly, but also think about nature-based solutions to climate, and increasingly thinking about that people corner of the triangle.
“That's one of the more challenging ones for us to think about. Often if you work on just climate and nature, you're in danger of disempowering and having unintended impacts on some of the most vulnerable people in the world. So at Google Deepmind we’re looking at how we address climate and nature without negatively affecting indigenous communities, for example.”
Robert sees AI as a key accelerator of sustainable development, and an exciting opportunity – when applied productively.
“Fugro is continuously investigating how we can use AI and further technologies. One of the biggest challenges we have is acquiring as much information from the data that we can acquire in one go, because we want to reduce our impact by reducing the times we measure.”
Drew agrees that AI can be a tool for efficiency, but that developers need to consider the possible implications. It is no secret that AI has complicated ethics in terms of its energy intensity and extensive applications, and when considering the importance, and sometimes fragility, of biodiversity.
“We need to be very cognizant of all of the potential unintended consequences of using AI, the energy and other resource use like the mining of the metals and materials,” Drew explains. “There are also potential safety, ethical implications, socio economic implications – we take it very seriously.”
The message is clear – technology has a place in the future of sustainability, and the opportunities that combining geodata and AI for the benefit and longevity of crucial biodiversity are extensive. It is, however, in the hands of governments, companies and organisations, as well as leaders like Google Deepmind and Fugro, to ensure that the developments are ethical, well-positioned and successful.
To see the full article, READ HERE.
******
Make sure you check out the latest edition of Sustainability Magazine and also sign up to our global conference series - Sustainability LIVE 2024
******
Sustainability Magazine is a BizClik brand
