Deep Learning Models for Predicting Wildfires from Historical Remote-Sensing Data

Identifying regions that have high likelihood for wildfires is a key component of land and forestry management and disaster preparedness. We create a data set by aggregating nearly a decade of remote-sensing data and historical fire records to predict wildfires. This prediction problem is framed as three machine learning tasks. Results are compared and analyzed for four different deep learning models to estimate wildfire likelihood. The results demonstrate that deep learning models can successfully identify areas of high fire likelihood using aggregated data about vegetation, weather, and topography with an AUC of 83%.

[1]  Thomas A. Hennig,et al.  The Shuttle Radar Topography Mission , 2001, Digital Earth Moving.

[2]  J. Abatzoglou.  Development of gridded surface meteorological data for ecological applications and modelling , 2013 .

[3]  David E. Rupp,et al.  Seasonal Climate Variability and Change in the Pacific Northwest of the United States , 2014 .

[4]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[5]  Dit-Yan Yeung,et al.  Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting , 2015, NIPS.

[6]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  J. Randerson,et al.  Global fire emissions estimates during 1997–2016 , 2017 .

[8]  Jaakko Kukkonen,et al.  Mortality due to Vegetation Fire–Originated PM2.5 Exposure in Europe—Assessment for the Years 2005 and 2008 , 2016, Environmental health perspectives.

[9]  Sean A. Parks,et al.  High-severity fire: evaluating its key drivers and mapping its probability across western US forests , 2018 .

[10]  Nathaniel P. Robinson,et al.  Mean Composite Fire Severity Metrics Computed with Google Earth Engine Offer Improved Accuracy and Expanded Mapping Potential , 2018, Remote. Sens..

[11]  Jigar Doshi,et al.  FireNet: Real-time Segmentation of Fire Perimeter from Aerial Video , 2019, ArXiv.

[12]  J. San-Miguel-Ayanz,et al.  A global wildfire dataset for the analysis of fire regimes and fire behaviour , 2019, Scientific Data.

[13]  Younes Oulad Sayad,et al.  Predictive modeling of wildfires: A new dataset and machine learning approach , 2019, Fire Safety Journal.

[14]  Jason Hickey,et al.  Machine Learning for Precipitation Nowcasting from Radar Images , 2019, ArXiv.

[15]  Yarin Gal,et al.  Machine Learning for Generalizable Prediction of Flood Susceptibility , 2019, ArXiv.

[16]  Padhraic Smyth,et al.  Machine learning to predict final fire size at the time of ignition , 2019, International journal of wildland fire.

[17]  Soukayna Mouatadid,et al.  WeatherBench: A Benchmark Data Set for Data‐Driven Weather Forecasting , 2020, Journal of Advances in Modeling Earth Systems.