Satellite images, when processed by artificial intelligence, could predict dangerous wildfires.
Traditional methods of predicting wildfires rely upon how much dry vegetation is present in an area, or in other words, the “fuel” that can trigger and sustain a wildfire. Scientists also look at how dry the landscape is. These methods are chancy at best, and difficult to assess quickly enough across large areas. However, the analysis of satellite images with the help of artificial intelligence offers hope for a more reliable and faster prediction of wildfires. (weather.com)
New study
“Over the last about 50 years, fire science has been based on an assumption that if we know how dry the weather is, we can tell how dry the forests are,” said Krishna Rao, one of the study’s authors and a doctoral student in Earth system science at Stanford University.
Using a different process, researchers used data from the European Space Agency’s Sentinel-1 satellites. This covered three years of moisture data from 239 sites across 12 West Coast US states that were vulnerable to wildfires.
The researchers trained an artificial intelligence system to recognize patterns in the data that would foretell a higher risk of wildfires.
They then compared the outputs with the latest outlook from the National Interagency Fire Centre. The researchers found that much the same areas are vulnerable to above-normal wildfire activity in the coming summer.
Climate change the culprit
“One of our big breakthroughs was to look at a newer set of satellites that are using much longer wavelengths, which allows the observations to be sensitive to water much deeper into the forest canopy and be directly representative of the fuel moisture content,” said Alexandra Konings, senior author on the study and ecohydrologist at Stanford University.
Additional factors are fluctuating weather patterns and climate change such as drought. As a result, plants are adapting to the new realities, and adversely affecting the conventional methods to predict wildfires.
However, with advanced imaging and sophisticated artificial intelligence, predictive models can quickly warn of deadly wildfires.
People can, therefore, move to safety sufficiently in advance of a calamity.
Related Story: AI Can Now Predict Lightning Strikes