AI System Predicts How to Prevent Wildfires
Aalto University (Finland)
September 9, 2022
A neural network model developed by researchers at Finland's Aalto University can predict the effectiveness of different land management strategies in preventing wildfires. Their analysis concentrated on Borneo's Central Kalimantan province, and the model predicted the distribution of peatland fires from measurements captured before each fire season from 2002 through 2019. The researchers analyzed 31 factors, then forecast the likelihood of a peatland fire at each spot on the map, resulting in an annual expected fire distribution that was correct 80% to 95% of the time. They then simulated the effects of different land management strategies on such fires, and found that converting shrubland and scrubland into swamp forests would cut fire incidence in half.
No comments:
Post a Comment