Mitigating the Risk of Wildfires with Accelerated Analytics

Climate change has already had a major impact across the United States. The Biden administration recently unveiled a 10-year plan to spend billions of dollars to combat destructive wildfires on roughly 50 million acres of land. Both operationally and to inform public policy, we need physical spread models and the assessment models built on them. Why? Because size matters. Conventional models treat all people as identical and fail to map areas much smaller than a census block.

During this webinar, learn how modern data science techniques and GPU-accelerated analytics can be combined with better data to improve outcomes. Using California as a case study, analysts start with disaggregate data from a Microsoft ML model, providing a building footprint for each building in the state (10 million in total). Then they consider forest structure data from the California Forest Observatory, also generated using ML models. At 10m/pixel, this amounts to 400m pixels per time slice (4) per variable (5).

To concentrate on wildfire risk for households, the fire science concept of “defensible space” can buffer buildings to extract surrounding forest conditions. Over the last four years, California has experienced some devastating fires, and their perimeters are available as open data. Learn how analysts resample census demographic data at the household level to estimate the total population and subsets of that population with special needs.

What do we learn from this data?

• We deploy unsupervised learning techniques and in particular cluster analysis. When applied to the biophysical data, this supports the classification of the ‘defensible space’ surrounding millions of buildings. When we add the social and fire history data, we can further characterize the relationships between people, landscape management, and outcomes.
• When we apply supervised techniques such as Random Forests, we can explore the factors correlated with particular outcomes.