Published works

Wildfire is possibly the most widespread natural disturbance globally, as its spatial and temporal patterns shape vegetation assemblages throughout the world. This is especially true in Australia, where the dominant trees, Eucalyptus spp., are among the most well-adapted to fire on the planet. To protect communities in such a fire-prone landscape, we need to be able to predict fire behaviour and manage fuels in a fashion that reduces risk. But to do this effectively, we must understand the natural fire regimes that occur in Australian ecosystems. One Australian ecosystem for which fire regimes are poorly understood is the tall wet Eucalyptus forest (TWEF). These globally unique forests support a mix of flammable and fire-sensitive vegetation and are among the world’s most carbon-dense forests, so understanding their flammability is critical.

This thesis attempts to describe the fire regime TWEF. It does so in a fashion that can inform fuels management and fire behaviour prediction, with a particular focus on the island of Tasmania. First, I used fire behaviour model simulations to analyse the effectiveness of fuel treatment on fire behaviour in different Tasmanian vegetation, which allowed me to contextualise fire risk across the island’s diverse vegetation types. I then described the fire regime, namely the expected frequency and severity of wildfires, of TWEF. I did this using a combination of fuels, microclimate, and weather data, along with fire behaviour modelling, to estimate potential fire severities in these forests. I first investigated what drives fire severity across a continental-scale macroecological gradient of forests in an ‘early-mature’ successional stage. However, flammability of older forests can be different from that of younger forests, so I investigated how the flammability of TWEF changes as a stand develops from regrowth to old-growth forest. For the latter exercise I focused, on Tasmanian TWEF.

Fire behaviour model simulations indicated that an impractical extent of landscape-scale prescribed burning would be needed for the treatment to be effective, and that this is especially true for TWEF. The results also suggested that Tasmanian TWEF are capable of sustaining among the most intense fires on the planet. I suggest that prescribed burning needs to be conducted at local, targeted scales, and that alternatives to prescribed burning need to be investigated for Tasmanian TWEF. Analysis of the continent-wide dataset of fuels and microclimate in early-mature TWEF indicated both low-and high-severity fires play an important role, but that high-severity fire is much more likely in TWEF in the hotter and drier regions of the continent. Further, similar analysis of a chronosequence of TWEF stands in Tasmania indicated that high-severity fires become more likely in younger TWEF. Importantly, the results highlighted that the TWEF understorey (or elevated fuel layer) is the most important fuel layer from a fire danger perspective. Hence, I suggest that management needs to mimic mixed-severity fires to maintain the structural heterogeneity and microclimate that makes these forests fire resistant.

Results from fire behaviour model simulations and from measuring fire severity in eight burnt field sites also indicated that current operational models over-predict flame height and fireline intensity in TWEF, likely due to an over-simplistic representation of forest structure. As a result, I performed a detailed review of the three major current operational fire behaviour models, along with a next-generation physics-based model, and documented their shortcomings. I then investigated how applicable these different models were in TWEF. Despite its importance, the live understorey is composed of many fire-sensitive species whose flammability is poorly understood. Hence, I propose that small-scale burning studies combined with physical simulation of wildfire are needed to quantify flammability and fire behaviour in forests composed of such species.