Climate change impacts on global wildfire ignitions by lightning and the safe management of landscape fuels

Lead Research Organisation: University of East Anglia
Department Name: Environmental Sciences

Abstract

2020 will be remembered not only for Covid-19, but also for its devastating wildfires. The year began with the Australian bushfires, which caused 34 deaths and over US$70 billion of damages. The wildfires that are currently burning across the Western US look set to be the costliest in US history, with over 30 people already killed directly. The wildfires of 2020 join a growing list of extreme wildfires seen in Mediterranean Europe, North America, Australia, Siberia and the Arctic in the past decade.

Wildfires are strongly tied to climatic droughts, which enhance the flammability of vegetation. As drought frequency is projected to rise in future, there are serious concerns that the fires seen in recent years are a glimpse into a more fire-prone future. Lightning strikes are the dominant cause of wildfire in many regions and, for example, ignited many of the recent Australian and US fire complexes.

It is critical that we understand the drivers of wildfire, build capacity to predict their future likelihood, and take steps to mitigate their impacts. Our current understanding of fires at the global scale is built around satellite observations. However, these observations are insufficient to disentangle the diverse drivers of fire; they see only patterns. Satellite observations provide a mixed signal of many different types of fire, including wildfires but also a range of fires under human control (e.g. agricultural fires and deforestation fires), meaning that observations of wildfire are 'contaminated' with other fire types. Critically, this obscures trends in wildfire activity and compromises our understanding of climate impacts on wildfire activity.

The proposed project will create the first global capacity to isolate lightning-ignited wildfires from satellite observations. It will use new observations of lightning strikes from ground- and satellite-based lightning sensors to 'decontaminate' satellite observations and introduce a global dataset of lightning-ignited wildfire activity. The new dataset will be used to make key advances in the understanding of wildfires and their relationship with climate. The project will assess how wildfire activity has changed in recent decades, and it will specifically determine the climatic conditions under which lightning fires occur. This new understanding of fire drivers will be built into the UK Earth System model and used to predict the impact of climate change on wildfire activity in the future century.

The new capacity to observe wildfires will also enable a major advance in the estimation of deforestation fire emissions, specifically in Amazonia which accounts around 40% of global emissions due to land use change. Deforestation fire emissions contribute to an increase in atmospheric concentrations of CO2 and contribute to climate change. However, emissions from Amazonian deforestation fires are known to be overestimated because emissions from lightning-ignited wildfires are undesirably included in the estimates. The new record of lightning-ignited wildfires developed in this study will be used to correct the emissions estimates and discount wildfires that occur as part of a natural disturbance-recovery cycle in the region.

Finally, this project will evaluate our future capacity to manage the threats of wildfires in a changing climate using conventional approaches to forest fuel management. It is common in some regions (e.g. Australia and the western US) to manage forest fuel stocks by burning off the most flammable fuels on the forest floor. However, this practice can only be applied safely during cool, moist, wind-free weather that occur in an annual 'window' of opportunity. It is feared that this window will narrow in future due to climate change. In this project, climate model outputs will be used to predict change in the window to 2100, informing forest management agencies of their future challenges and resource needs.

Publications

10 25 50