Prescribed Fire Modeling using Knowledge-Guided Machine Learning for Land Management.

Somya Sharma, Kelly Lindsay, Neel Chatterjee, Rohan Prashant Patil, Ilkay Altintas, Michael Steinbach, Daniel Giron, Mai H Nguyen, Vipin Kumar

Abstract:

In recent years, the increasing threat of devastating wildfires
has underscored the need for effective prescribed fire man-
agement. Process-based computer simulations have tradi-
tionally been employed to plan prescribed fires for wildfire
prevention. However, even simplified process models are too
compute-intensive to be used for real-time decision-making.
Traditional ML methods used for fire modeling offer com-
putational speedup but struggle with physically inconsistent
predictions, biased predictions due to class imbalance, bi-
ased estimates for fire spread metrics (e.g., burned area, rate
of spread), and limited generalizability in out-of-distribution
wind conditions. This paper introduces a novel machine
learning (ML) framework that enables rapid emulation of
prescribed fires while addressing these concerns. To over-
come these challenges, the framework incorporates domain
knowledge in the form of physical constraints, a hierarchi-
cal modeling structure to capture the interdependence among
variables of interest, and also leverages pre-existing source
domain data to augment training data and learn the spread
of fire more effectively. Notably, improvement in fire met-
ric (e.g., burned area) estimates offered by our framework
makes it useful for fire managers, who often rely on these
estimates to make decisions about prescribed burn manage-
ment. Furthermore, our framework exhibits better general-
ization capabilities than the other ML-based fire modeling
methods across diverse wind conditions and ignition pat-
terns.