How AI is aiding Google in fighting fires

From 2001 to 2023, Kenya lost 2,320 hectares of tree cover from fires and 384,000 hectares from all other drivers of loss.

The year with the most tree cover loss due to fires during this period was 2022, with 190 ha lost to fires equivalent to 2.9 per cent of all tree cover loss for that year. With climate change, things could get worse.

This is why Google's latest wildfire boundary maps expansion to more countries including Kenya is especially timely.

The Silicon Valley company is expanding this tool to 15 countries across Europe and Africa: Andorra, Bosnia and Herzegovina, Croatia, Cyprus, France, Greece, Italy, Kenya, Monaco, Montenegro, Portugal, Rwanda, Slovenia, Spain and Turkiye.

You probably can't outrun a forest fire. They can travel at speeds of up to 22 kilometres per hour and are dangerously unpredictable. Staying one step ahead of a wildfire can mean the difference between life and death.

Google is now adding satellite data into its mapping facility to show the precise locations of wildfires in close to real-time.

Using AI, Google can show detailed wildfire boundary tracking in Search and Maps and through location-based push notifications. Anyone in the vicinity of a fire, and using Google Maps, can be alerted to the danger.

To ensure accuracy, Google's AI-powered wildfire tracking model is trained on multiple sources of data, including a variety of satellite imagery data sets. As one of the world's most powerful collectors of data, Google is using AI to help the world adapt to the impacts of climate change and the hazards it's causing.

Climate change is partly responsible for wildfire spread or is at the very least becoming a contributory factor.

However, to help maintain biodiversity, forest fires are part of the natural cycle of the woodland ecosystem. According to Nasa's Earth observation, fire kick starts the germination process of many seeds, it also "initiates critical natural processes by breaking down organic matter into soil nutrients."