27 November 2023

AI for Climate Change: Preventing and predicting wildfires 

By Kate Iida

As Earth’s temperature rises, wildfires are getting worse. Can artificial intelligence help prevent (and predict) them?

From Greece, Portugal and Spain, to Hawaii and Canada, devastating wildfires have taken their toll this year. Copernicus, the European Union’s Earth observation programme, reported that in Canada, fires have contributed to the country’s total of 410 megatonnes of carbon emissions over 2023, even before the year has ended. This is the highest level of annual carbon emissions ever recorded in Canada, and broke the country’s previous record established in 2014. Copernicus has said that these figures may even still continue to rise.

Wildfires are increasingly becoming one of the most important issues related to climate change. On the one hand, global warming leads to greater and more devastating wildfires, which cause stress and suffering for people, communities, and the natural environment. Beyond this, large scale wildfires contribute significantly to carbon emissions, in turn leading to greater global warming.  In order to adapt to the changes the Earth is experiencing as a result of climate change and to slow the rise of global carbon emissions, countries need to learn how to prevent wildfires and keep them under control.

How can AI help?

In some places bearing the brunt of wildfire damage, governments are already beginning to explore using emerging technologies such as artificial intelligence to detect fires when they have recently started, and even to predict which areas are most vulnerable to fires and which conditions could create a fire that could spread out of control.

In the sections below, I’ll explore some of these initiatives, looking at their successes, potential and also what some of their limitations could be. Several of these examples come from California, my home state, making this an especially personal issue to me.

Artificial intelligence for wildfire detection

California has faced more than its share of devastating wildfires in the past few years. The 2018 Camp Fire, which spread through Butte county in Northern California, remains the deadliest and most destructive fire in the state’s history, leading to the death of 85 people and the destruction of an entire community. Though 2023 has been considered a “light” year for wildfires, more than 4,700 wildfires broke out across the state over just this year. It’s as important as ever for California to remain vigilant.

California also has access to companies and people skilled in developing and using artificial intelligence, due to the state’s position as a leader in tech. This summer, multiple firefighting departments across the state have begun programmes to use AI technologies to help them detect fires early.

Cal Fire, the state’s firefighting department, is currently working with the University of San Diego’s AlertCalifornia system on a project to use AI to alert firefighters early about fires that have just begun. Currently, the firefighting department needs to wait until someone has called in about a fire to know that it has started. This means that if a fire has started in a particularly remote area, for example, it can have significant time to spread before firefighters even know what is happening.

The AI system, which CalFire piloted this summer, is looking to fix that. The system works by scanning the feeds of 1,039 high definition cameras that have been spread throughout different “strategic locations” across the state. The AI looks to detect “anomalies” in the images sent in by the cameras, such as evidence of smoke. When the system identifies something unexpected, it alerts local firefighting officials, who check the images to determine whether it’s something they should respond to.

This summer, the system was piloted in six of CalFire’s emergency command centres throughout the state. According to the Los Angeles Times, officials say that the system correctly identified 77 fires before any officials had received a call about them, allowing firefighters to respond early before the fire could potentially spread out of control. The system is now going to be expanded to be used in 21 of Cal Fire’s emergency command centres across the state.

Along with this initiative, several counties in California are also looking to implement their own AI systems to identify fires early. Santa Clara county, located in Northern California, is looking to implement AI sensors to identify fires, NBC Bay Area reported. These sensors would analyse air temperature and particles in the air, specifically gas levels, matter, and heat, with the intention of locating fires when they’ve just begun. The county intends to place the sensors near the Santa Clara County watershed, as this is an area where wildfires could cause significant disruption if it were to damage the piping system that sends water to the county’s residents. The county of Oakland has already implemented similar sensors, the article reports.

These initiatives seem promising, but they haven’t been in use for very long. It will take more time, and the expansion of initiatives in greater numbers across the state, to see whether they’re able to truly make a difference in reducing the incidence and severity of California’s fires.

Artificial intelligence for wildfire prevention

What would happen if governments could know a deadly fire was about to occur, even before it had started? This is the question that researchers are asking as part of two international initiatives looking to use artificial intelligence technologies to predict wildfires before they occur, to help governments make more informed decisions about what prevention strategies to use.

FireAId, an initiative of the World Economic Forum, is looking to use AI trained on meteorological and static datasets to create a risk map of a given area. The risk map categorises different locations by “ignition,” or the probability of a wildfire starting in that area, and “severity” or how dangerous the fire could be. The initiative has been piloted already in Turkey, with a focus on the South Aegean and West Mediterranean regions. In recent years, Turkey has experienced multiple wildfires, in particular in 2021, when more than fifty fires blazed at once, leading to the deaths of multiple people in tourist regions near the Aegean Sea.

The FireAId risk map has been used to create a model to help local departments determine how to allocate resources in the most effective way. Now that it has passed its pilot phase, it is going to be expanded to be used across the whole country. The expanded version will also include forecasting models and a dynamic map that people using the model can use to explore relevant statistics for the region. FireAId is also looking into the possibility of developing digital twins of the fire prone areas, which would allow firefighters to explore the most likely progress of the fire’s path and the effectiveness of different proposed responses.

Despite the initial successes of this project, firefighting responses in Turkey still face challenges. In many places, firefighters are working with outdated equipment and on small budgets. The World Economic Forum expresses that in order to make the most of the predictions offered by the FireAID initiative, firefighters need more funding and more up to date equipment to give them the best chance to respond effectively.

Another fire prevention initiative comes from the University of Aalto in Finland, where researchers have developed an AI model intended to predict wildfires in Indonesia, according to Dezeen. The model is currently focussed around Indonesia’s Central Kalimantan province on Borneo and it uses climate data, including information about land cover, vegetation, and ongoing droughts, along with information about historical fire patterns to make its predictions. It also makes predictions about how different interventions could prevent or reduce the incidences, severity or intensity of future wildfires.

The intention of the researchers is that the model could be used by local policymakers to help  understand what interventions, such as changes around forest management, would have the biggest impact in reducing wildfires in the region.

Since it was developed, the AI model has predicted approximately ninety per-cent of the fires that occurred in that region in Indonesia.The creators of the model are hoping to test it next in the Mediterranean, where they could see whether it works well in another environment.

What next?

Wildfires are one of the most pressing problems of climate change, and as global warming increases they will become even more frequent.  The initiatives explored in this blog have potential, and have shown in many cases to work, though there are still ways in which the accuracy of the predictive models can be improved. It’s also important to mention, however, that every climate change solution based on AI technologies has its own environmental consequences, which I’ve explored in a previous blog post about big data and climate change. Using AI technologies to help prevent and predict wildfires can be helpful, but at the moment, these initiatives are still very early in their design process, and have only been used in a small number of places. As the number and strength of wildfires increase in the future, it’s likely that AI technologies will be used even more as another tool for firefighters, to help them adapt and respond quickly when, and even before, they begin.


More insights

21 April 2017

Why Government is ready for AI

12 July 2017

Five levels of AI in public service

26 July 2017

Making it personal: civil service and morality

10 August 2017

AI: Is a robot assistant going to steal your job?

19 September 2017

AI and legitimacy: government in the age of the machine

06 October 2017

More Than The Trees Are Worth? Intangibles, Decision-Making, and the Meares Island Logging Conflict

16 October 2017

The UK Government’s AI review: what’s missing?

23 October 2017

Why unconference? #Reimagine2017

03 November 2017

AI: the ultimate intern

09 November 2017

Motherboard knows best?

23 November 2017

Beyond driverless cars: our take on the UK’s Autumn Budget 2017

05 December 2017

Why Black people don’t start businesses (and how more inclusive innovation could make a difference)

06 December 2017

“The things that make me interesting cannot be digitised”: leadership lessons from the Drucker Forum

23 January 2018

Want to get serious about artificial intelligence? You’ll need an AI strategy

15 February 2018

Economic disruption and runaway AI: what can governments do?

26 April 2018

Ranking governments on AI – it’s time to act

08 May 2018

AI in the UK: are we ‘ready, willing and able’?

24 May 2018

Mexico leads Latin America as one of the first ten countries in the world to launch an artificial intelligence strategy

05 July 2018

Beyond borders: talking at TEDxLondon

13 July 2018

Is the UK ready, willing and able for AI? The Government responds to the Lords’ report

17 July 2018

Suspending or shaping the AI policy frontier: has Germany become part of the AI strategy fallacy?

27 July 2018

From open data to artificial intelligence: the next frontier in anti-corruption

01 August 2018

Why every city needs to take action on AI

09 August 2018

When good intentions go bad: the role of technology in terrorist content online

26 September 2018

Actions speak louder than words: the role of technology in combating terrorist content online

08 February 2019

More than STEM: how teaching human specialties will help prepare kids for AI

02 May 2019

Should we be scared of artificial intelligence?

04 June 2019

Ethics and AI: a crash course

25 July 2019

Dear Boris

01 August 2019

AI: more than human?

06 August 2019

Towards Synthetic Reality: When DeepFakes meet AR/VR

19 September 2019

Predictive Analytics, Public Services and Poverty

10 January 2020

To tackle regional inequality, AI strategies need to go local

20 April 2020

Workshops in an age of COVID and lockdown

10 September 2020

Will automation accelerate what coronavirus started?

10 September 2020

Promoting gender equality and social inclusion through public procurement

21 September 2020

The Social Dilemma: A failed attempt to land a punch on Big Tech

20 October 2020

Data and Power: AI and Development in the Global South

23 December 2020

The ‘Creepiness Test’: When should we worry that AI is making decisions for us?

13 June 2022

Data promises to support climate action. Is it a double-edged sword?

30 September 2022

Towards a human-centred vision for public services: Human-Centred Public Services Index

06 October 2022

Why You Should Know and Care About Algorithmic Transparency

26 October 2022

Harnessing data for the public good: What can governments do?

09 December 2022

Behind the scenes of the Government AI Readiness Index

06 February 2023

Reflections on the Intel® AI for Youth Program

01 May 2023

Canada’s AI Policy: Leading the way in ethics, innovation, and talent

15 May 2023

Day in the life series: Giulia, Consultant

15 May 2023

Day in the life series: Emma, Consultant

17 May 2023

Day in the life series: Kirsty, Head of Programmes

18 May 2023

Day in the life series: Sully, Partnerships Associate/Consultant

19 May 2023

LLMs in Government: Brainstorming Applications

23 May 2023

Bahrain: Becoming a regional R&D Hub

30 May 2023

Driving AI adoption in the public sector: Uruguay’s efforts on capacity-building, trust, and AI ethics

07 June 2023

Jordan’s AI policy journey: Bridging vision and implementation

12 June 2023

Response to the UK’s Global Summit on AI Safety

20 June 2023

 Unlocking the economic potential of AI: Tajikistan’s plans to become more AI-ready

11 July 2023

Government transparency and anti-corruption standards: Reflections from the EITI Global Conference in Dakar, Senegal

31 August 2023

What is quantum technology and why should policymakers care about it?

21 September 2023

Practical tools for designers in government looking to avoid ethical AI nightmares

23 October 2023

Collective Intelligence: exploring ‘wicked problems’ in National Security

23 October 2023

Exploring the concepts of digital twin, digital shadow, and digital model

30 October 2023

How to hire privileged white men

09 November 2023

Inclusive consensus building: Reflections from day 4 of AI Fringe

13 November 2023

AI for Climate Change: Can AI help us improve our home’s energy efficiency?

14 November 2023

Navigating the AI summit boom: Initial reflections

20 November 2023

AI for Climate Change: Improving home energy efficiency by retrofitting

24 November 2023

Will AI kill us all?

28 November 2023

Service Design in Government 2023: conference reflections

04 December 2023

AI for Climate Change: Using artificial and indigenous Intelligence to fight climate change

06 December 2023

Release: 2023 Government AI Readiness Index reveals which governments are most prepared to use AI

11 December 2023

AI for Climate Change: AI for flood adaptation plans and disaster relief

18 December 2023

AI for Climate Change: Managing floods using AI Early Warning Systems