How AI can help fight climate change

We live in a world where climate change is already a reality.  

The polar ice caps are melting, sea levels are rising, and we are seeing more storms, torrential rains, extreme heat waves and droughts. The consequences for nature are severe: water availability is decreasing, marine ecosystems are being disrupted and biodiversity is declining.  

Scientists around the world are working to reverse these trends. At the same time, they are developing solutions to meet the challenges we already face - and those that lie ahead.

Increasing biodiversity can be part of the solution 

Olof Mogren is one of five research leaders at the Centre for Applied AI at RISE.

"Generally speaking, AI is a collection of techniques whose greatest strength is learning to recognise complex relationships through training on large amounts of data. This is a skill that is perfectly suited to work on climate change. By enabling AI to perform optimisation, analysis and even design, it can become part of the toolbox we need to tackle this fateful issue."

Olof Mogren believes that the rapid developments in the field are creating more opportunities every day.

"In principle, data should be collected as completely as possible with as little distortion as possible, but in practice this does not always work. One of the great things about modern AI modelling is that it is very flexible. We can put together different types of modules and different sources of data that together form a unit that gives good results."

Olof Mogren's research group works extensively on climate-related issues - including how AI can benefit biodiversity.

"Biodiversity loss is often a symptom of climate change. So measures to improve one will often have a positive effect on the other," he says, citing two examples: wetland restoration and revegetation, where whole areas are restored to their original state with little or no human intervention." 

AI provides a quick snapshot of the state of biodiversity 

When it comes to biodiversity, Olof Mogren's team uses what they call advanced AI-based data analysis. The idea is to quantify and track where species are, how they feel and how they behave.

"One of the things we do is what we call soundscape analysis. This means that we work with sound boxes that record what is happening in the environment. With the help of the AI we are developing, we are making the boxes intelligent so that you can draw conclusions from what you hear."

"Today," says Olof Mogren, "many biologists do the work of analysing the material manually or semi-automatically, which is very time-consuming."

"But with the help of AI, we can more quickly and better identify which species are heard on such a recording. Otherwise, it is a time-consuming task for an expert to listen to months of recordings."

Many of the species heard on recordings can be seen as markers of biodiversity.  

"We can use artificial intelligence to quantify the species in an area. This gives us an idea of how the biological system on the site is doing."

Analyses inform decisions 

In addition to in-depth analysis, AI can also be used to remotely analyse satellite data. This is particularly useful when researchers are looking at how different habitats are changing and what action can be taken in the ongoing process known as the sixth mass extinction.  

"Satellite images provide a general overview, but today they can also be quite high-resolution. Using advanced algorithms, we can analyse which areas are suitable for which species."

All these analyses can be used to make decisions about what action is needed.

"For example, we can use the analyses to decide which areas are best suited for rewilding."

The analyses can also inform decisions on how to create new or restore wetlands that have the potential to sequester carbon and thus become a carbon sink.

"It is not necessarily the case that high-tech in the form of AI will be at the heart of climate change solutions. But I think we can use high-tech to enhance or improve low-tech solutions such as reforestation or wetland restoration."

Businesses can use AI in their environmental and climate strategies 

RISE is working with several stakeholders on AI tools for climate. These include SMHI, the Swedish University of Agricultural Sciences, Lund University and Uppsala University, as well as several municipalities and county councils. But Olof Mogren wants more to join in.

"Municipalities and other organisations are welcome, but so are companies. Most companies could benefit from this type of technology in their environmental and climate strategy. Either to increase their competitiveness or to comply with existing and future legislation."

Olof Mogren believes that all actors should test AI as a way of leveraging their data collection to make smarter decisions for their business and for the world.

"By contacting us at RISE, you can get help in looking at the data and discussing what predictive qualities it has and what kind of problems the analysed data can help solve."