Deep Learning Methods for Daily Wildfire Danger Forecasting
Abstract
Wildfire forecasting is of paramount importance for disaster risk reduction and environmental sustainability. We approach daily fire danger prediction as a machine learning task, using historical Earth observation data from the last decade to predict next-day's fire danger. To that end, we collect, pre-process and harmonize an open-access datacube, featuring a set of covariates that jointly affect the fire occurrence and spread, such as weather conditions, satellite-derived products, topography features and variables related to human activity. We implement a variety of Deep Learning (DL) models to capture the spatial, temporal or spatio-temporal context and compare them against a Random Forest (RF) baseline. We find that either spatial or temporal context is enough to surpass the RF, while a ConvLSTM that exploits the spatio-temporal context performs best with a test Area Under the Receiver Operating Characteristic of 0.926. Our DL-based proof-of-concept provides national-scale daily fire danger maps at a much higher spatial resolution than existing operational solutions.
Cite
@article{arxiv.2111.02736,
title = {Deep Learning Methods for Daily Wildfire Danger Forecasting},
author = {Ioannis Prapas and Spyros Kondylatos and Ioannis Papoutsis and Gustau Camps-Valls and Michele Ronco and Miguel-Ángel Fernández-Torres and Maria Piles Guillem and Nuno Carvalhais},
journal= {arXiv preprint arXiv:2111.02736},
year = {2021}
}
Comments
Accepted to the workshop on Artificial Intelligence for Humanitarian Assistance and Disaster Response at the 35th Conference on Neural Information Processing Systems (NeurIPS 2021)