Modelling bushfire severity and predicting future trends in Australia using remote sensing and machine learning

Abstract

Bushfires are one of the major natural disasters that cause huge losses to livelihoods and the environment. Understanding and analysing the severity of bushfires is crucial for effective management and mitigation strategies, helping to prevent the extensive damage and loss caused by these natural disasters. This study presents an in-depth analysis of bushfire severity in Australia over the last twelve years, combining remote sensing data and machine learning techniques to predict future fire trends. By utilizing Landsat imagery and integrating spectral indices like NDVI, NBR, and Burn Index, along with topographical and climatic factors, we developed a robust predictive model using XGBoost. The model achieved high accuracy, 86.13%, demonstrating its effectiveness in predicting fire severity across diverse Australian ecosystems. By analysing historical trends and integrating factors such as population density and vegetation cover, we identify areas at high risk of future severe bushfires. Additionally, this research identifies key regions at risk, providing data-driven recommendations for targeted firefighting efforts. The findings contribute valuable insights into fire management strategies, enhancing resilience to future fire events in Australia. • The XGBoost model predicts bushfire severity with 86.13% R 2 due to data cleansing. • Identifying high-risk fire regions helps allocate resources based on trends and vegetation. • Mapping 12 years of data shows recurring high-severity fires in north and east Australia. • Population density and vegetation data guide firefighting efforts in high-risk areas. • The study highlights real-time data use for better fire prediction and resource planning.

Topics & Concepts