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 openaccess 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. † Institute for Astronomy, Astrophysics, Space Applications and Remote Sensing 1 https://cwfis.cfs.nrcan.gc.ca/background/summary/fdr 2 https://www.afac.com.au/initiative/afdrs 35th Conference on Neural Information Processing Systems (NeurIPS 2021),

Analysis of Trends in the Distance of Wildfires from Built-Up Areas in Spain and California (USA): 2007–2015

“…The scientific community [100] has been discussing the selection of a spatial scale for studying wildfire forecasting. The spatial scale of distribution-based approaches varies from fine-scale grids, which are typically 1 km x 1 km or smaller [101][102][103][104], to larger scales of approximately 10 km × 10 km [105][106][107], multiscalar [108,109] or by using computerized and artificial intelligence techniques [110,111]. Initially, two scale levels were selected: level 8 (7774 km s 2 ) and level 9 (2591 km s 2 ).…”

Wildfire Danger Prediction and Understanding With Deep Learning

“…Predicting fire danger with ML is not straightforward and poses some challenges (Prapas et al, 2021), which are important to address:…”

Deep Learning Approaches for Wildland Fires Using Satellite Remote Sensing Data: Detection, Mapping, and Prediction

“…A higher accuracy of 95.81% was achieved, outperforming four benchmark models that are multilayer perceptron neural networks, random forests, kernel logistic regression, and SVM. Prapas et al [117] proposed a DL method, named ConvLSTM, for forest fire danger forecasting in the regions of Greece. The Datacube dataset [118] was used in training and testing this model.…”