10th International Aerosol Conference
September 2 - September 7, 2018
America's Center Convention Complex
St. Louis, Missouri, USA

Abstract View

Predicting Wildfire in United States Using Artificial Neural Network Technique

KAIYU CHEN, Hao Guo, Hongliang Zhang, Louisiana State University

Abstract Number: 50
Working Group: Control and Mitigation

Abstract
Wildfire is the largest source of air pollutants such as particulate matters and toxins, which greatly changes landscape with impacts on human health, large scale climate changes, social economy and ecosystem in the U.S. In this study, Artificial Neural Network (ANN) is applied to simulate the risk of wildfire in the U.S. based on the date recorded from 2000 to 2016. Relationship between wildfire activities, represented by fire occurrence and its brightness and fire radiative power (FRP), with climate factors including temperature (TEM), relative humidity (RH), wind speed (WS) and wind direction (WD) was simulated. Trained by using observation date, our model is reliable to show correlation between climate factors and wildfire activities. Via this model, we are able to estimate fire by detecting climate conditions from closest meteorology station. Results from this study could provide information for wildfire prediction, prevention and controlling management.