Modelling ignition probability and wildfire emissions in the boreal forest under climate change
Mr GAO Cong
(Supervisor: Prof Jinbao Li)
Wildfires are an important disturbance factor driving the biodiversity and carbon balance of forest ecosystems. Boreal forests, rich in forest resources and carbon stocks, play an important role in maintaining biodiversity and resisting climate change. With rapid warming at northern high latitudes, boreal forests are facing an unprecedented wildfire crisis, with an increase in the frequency, intensity, and burned area of wildfires. Predicting and preventing boreal wildfires has become a major challenge for human response to climate change. However, previous prediction models based on traditional algorithms have low accuracy, the domain drivers of wildfire activity in the boreal forest remain to be further explored, and the projections of wildfire emissions in the boreal forest are lacking. To address these concerns, this study proposes to develop models of wildfire activity based on advanced machine learning algorithms; introduce a novel interpretation method to identify the driving factors responsible for ignition and emission and explore their marginal effects; project the historical and future boreal forest wildfire activity and generate the wildfire susceptibility maps under different scenarios. This study aims to enhance the comprehension and assessment of boreal wildfire risks and emissions under climate change, and to support timely wildfire prevention and carbon management policies.