Enabling Real-Time Training of a Wildfire-to-Smoke Map with Multilinear Operators
Abstract
Wildfires are a major producer of fine particulate matter, impacting human health and the electrical grid. Accurately forecasting smoke impacts over long time scales incorporates fuel treatment strategies, natural fuel succession, and stochastic events like lightning strikes. However, predicting smoke for each fuel distribution with a forward simulation of a coupled fire-atmosphere model is computationally infeasible. Moreover, relatively simple fire models are tractable to run in many long-time scenarios but do not capture smoke transport. We use data-driven multilinear operators to predict a smoke concentration field from knowledge of the time since ignition for two quantities of interest: aerosol optical depth and smoke detection. Our method first computes the principal components of time-since-ignition and smoke concentration fields and then learns a map from powers of the input coefficients to the output coefficients. We apply our learned operator to smoke prediction in the Upper Rio Grande Watershed. After collecting training data, learning the approximation weights on a CPU takes less than 30 seconds, and each forward call takes less than 1 ms. On a proxy for aerosol optical depth, we obtain equal accuracy to Monte Carlo sampling with fewer than half as many coupled model calls. For smoke detection, we obtain an intersection-over-union (IoU) of 0.64 and an area under the receiver operating characteristic curve (AUC) of 0.95 on holdout data. Our method is significantly more accurate than the most similar published smoke classifier, which obtains an IoU and AUC of 0.15 and 0.61, respectively, on a 2015 bushfire in Australia.
Versions
Related
- The July 2026 North American wildfire smoke event was three times deadlier due to climate change
- Long range transport of Canadian wildfire smoke to Europe in 2023: aerosol properties and spectral features of smoke particles
- TRAILS – a novel framework for time-height-resolved attribution of long-range transported wildfire smoke
- Geographic and thematic gaps in wildfire resilience research
- Wildfire risk for main vegetation units in a biodiversity hotspot: modelling approach in New-Caledonia, South Pacific.
BibTeX
@misc{morrow2026enabling,
title = {{Enabling Real-Time Training of a Wildfire-to-Smoke Map with Multilinear Operators}},
author = {Zachary Morrow and Joseph Crockett and John D. Jakeman and Dan J. Krofcheck},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.04164},
eprint = {2605.04164},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.04164},
}
RIS
TY - UNPB
TI - Enabling Real-Time Training of a Wildfire-to-Smoke Map with Multilinear Operators
AU - Zachary Morrow
AU - Joseph Crockett
AU - John D. Jakeman
AU - Dan J. Krofcheck
PY - 2026
DA - 2026/09/09
PB - arXiv
DO - 10.48550/arxiv.2605.04164
UR - https://arxiv.org/abs/2605.04164
AB - Wildfires are a major producer of fine particulate matter, impacting human health and the electrical grid. Accurately forecasting smoke impacts over long time scales incorporates fuel treatment strategies, natural fuel succession, and stochastic events like lightning strikes. However, predicting smoke for each fuel distribution with a forward simulation of a coupled fire-atmosphere model is computationally infeasible. Moreover, relatively simple fire models are tractable to run in many long-time scenarios but do not capture smoke transport. We use data-driven multilinear operators to predict a smoke concentration field from knowledge of the time since ignition for two quantities of interest: aerosol optical depth and smoke detection. Our method first computes the principal components of time-since-ignition and smoke concentration fields and then learns a map from powers of the input coefficients to the output coefficients. We apply our learned operator to smoke prediction in the Upper Rio Grande Watershed. After collecting training data, learning the approximation weights on a CPU takes less than 30 seconds, and each forward call takes less than 1 ms. On a proxy for aerosol optical depth, we obtain equal accuracy to Monte Carlo sampling with fewer than half as many coupled model calls. For smoke detection, we obtain an intersection-over-union (IoU) of 0.64 and an area under the receiver operating characteristic curve (AUC) of 0.95 on holdout data. Our method is significantly more accurate than the most similar published smoke classifier, which obtains an IoU and AUC of 0.15 and 0.61, respectively, on a 2015 bushfire in Australia.
ER -