Year: 2020
Type: dissertation
Abstract: Exacerbated by global change, wildfires are causing unprecedented impacts on ecosystems, human lives and infrastructure. Thus, there is a pressing need to better understand the drivers of wildfire in this changing world. I investigated interspecific patterns in plant leaf flammability in fire-prone vegetation of eastern Australia to determine the flammability dynamics of leaves given their essential role as fuels for wildfires. \nAt a landscape scale, I found that leaves of forest gully species had faster ignitibility and higher combustibility relative to woodland ridgetop species. These differences were driven by strong relationships between high flammability and both large leaf area and low leaf mass per area. For these plant communities, I explored relationships among leaf ignitibility, sustainability and combustibility and showed that faster ignitibility (higher flammability) was associated with short sustainability (lower flammability), and long sustainability (higher flammability) with higher combustibility (high flammability). Given the opposing relationship between ignitibility and sustainability, I established an overall leaf flammability index to assist in identifying low-leaf-flammability species at the wildland-urban interface. Close inspection of the flammability of ten species showed that increasing radiant energy, representing increases from low to moderate intensity wildfires, led to faster ignitability and a higher proportion of leaves flaming, such that species identity became much less important for understanding flammability dynamics. However, species identity remained particularly important for leaf sustainability and time to flaming, signalling that species differences in these flammability attributes may be helpful in understanding the dynamics of moderate intensity wildfires. Finally, I built predictive models of leaf ignitibility, sustainability and combustibility using a combination of radiant energy and leaf area, mass per area, and water content, and tested the accuracy of the models using a validation dataset. Including leaf traits in these models dramatically increased the ability to predict leaf flammability. One application of these models will be to estimate leaf flammability for large numbers of plant species using just their leaf traits, without the need for flammability experiments if they are not feasible. \nThis thesis provides novel insight into interspecific variation in plant leaf flammability. The findings of my research advance our ecological understanding of leaf flammability, with potential applications to wildfire modelling, the selection of low-leaf-flammability species at the wildland-urban interface, and prediction of leaf flammability for large numbers of species using leaf traits.
Author: Daniel W. Krix
Language: English
Topic: Fire effects on ecosystems
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
SDGs: Life on land, Climate action
Open Access status: green
OpenAlex ID: W3023557063