Deep Learning Methods for Daily Wildfire Danger Forecasting
This work addresses wildfire forecasting for disaster risk reduction, providing a proof-of-concept for national-scale daily maps at higher resolution than existing solutions, but it is incremental as it applies known deep learning methods to this domain.
The paper tackled daily wildfire danger forecasting by implementing various deep learning models using historical Earth observation data, achieving a best test AUC of 0.926 with a ConvLSTM model that outperformed a Random Forest baseline.
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.