Boundary-Aware Uncertainty Quantification for Wildfire Spread Prediction

TL;DR

This paper introduces a boundary-aware uncertainty quantification framework for wildfire spread prediction, emphasizing operational relevance within critical fire zones, and compares ensemble and student models using this new evaluation protocol.

Contribution

It proposes the Fire-Centered Evaluation Region (FCER) framework for spatially conditioned UQ assessment in wildfire modeling, enhancing boundary-sensitive evaluation methods.

Findings

The student model achieves comparable calibration to the ensemble.

FCER provides a more operationally relevant UQ evaluation.

Code is available at https://github.com/jonasvilhofunk/WildfireUQ-FCER.

Abstract

Reliable wildfire spread prediction is vital for risk-aware emergency planning, yet most deep learning models lack principled uncertainty quantification (UQ). Further, for boundary-sensitive cases like wildfire spread, evaluating models with global metrics alone is often insufficient. To shift the focus of UQ evaluation toward a more operationally relevant approach, the Fire-Centered Evaluation Region (FCER) framework is introduced as a spatially conditioned protocol to characterize UQ within critical fire zones. Using FCER, an Ensemble is compared against an distilled single-pass student model on the WildfireSpreadTS dataset. The student model demonstrates comparable calibration and complementary uncertainty ranking in boundary-relevant regimes. Code is available at https://github.com/jonasvilhofunk/WildfireUQ-FCER