Dense Forecasting of Wildfire Smoke Particulate Matter Using Sparsity Invariant Convolutional Neural Networks

TL;DR

This paper introduces a sparsity-invariant convolutional neural network that improves dense wildfire smoke PM2.5 forecasts by leveraging spatial biases and multitask learning, outperforming existing systems during recent wildfire seasons.

Contribution

The work presents a novel sparsity-invariant CNN architecture that effectively forecasts wildfire smoke particulate matter at high resolution, even with limited ground truth data.

Findings

Outperforms existing smoke forecasting systems in British Columbia

Predicts PM2.5 24 hours ahead at 10 km resolution

Generalizes well to realistic smoke dispersion patterns

Abstract

Accurate forecasts of fine particulate matter (PM 2.5) from wildfire smoke are crucial to safeguarding cardiopulmonary public health. Existing forecasting systems are trained on sparse and inaccurate ground truths, and do not take sufficient advantage of important spatial inductive biases. In this work, we present a convolutional neural network which preserves sparsity invariance throughout, and leverages multitask learning to perform dense forecasts of PM 2.5values. We demonstrate that our model outperforms two existing smoke forecasting systems during the 2018 and 2019 wildfire season in British Columbia, Canada, predicting PM 2.5 at a grid resolution of 10 km, 24 hours in advance with high fidelity. Most interestingly, our model also generalizes to meaningful smoke dispersion patterns despite training with irregularly distributed ground truth PM 2.5 values available in only 0.5% of grid cells.