Degree Name

Bachelor of Conservation Biology (Honours)


School of Earth, Atmospheric and Life Sciences


Owen Price


Fuel load is one of the primary determinants for fire behaviour in Australian forests. In South-Eastern Australia, ground and elevated fuel loads are generally considered highest at 15-20 years post-fire. Current methodology for predicting fuel load often relies on low resolution vegetation maps, simple time-since-fire relationships, and often incorrectly used ground-fuel models for elevated and canopy fuels. The combination of these prevents many modern fuel load models from revealing the fine-scale processes that truly drive fuel load accumulation. This study seeks to correct these issues by using high-resolution Light Detection and Ranging (LiDAR) alongside generalised additive models (GAM) and a variety of fire history and environmental variables to produce a nuanced model of post-fire vegetation recover. This study utilised aerial LiDAR point-cloud data to assess persistent, broad-scale changes in vegetation cover and structure after wildfire in dry sclerophyll forests of south-eastern Australia. The use of GAMSs allowed for the investigation of non-linear, polynomial relationships between vegetation cover and time-since-fire. Vegetation cover scores were produced at fine spatial scale (200 meters) for five dry sclerophyll forest classes across eastern New South Whales, Australia (Clarence, north coast, northern gorge, Sydney coastal, and Sydney hinterland dry sclerophyll forests). These scores were further classified by height into five strata: low shrubs (0.5-2 m), tall shrubs (2-5 m), low trees (5-10 m), medium trees (10-30 m), and tall trees (>30 m). The models produced in this study confirmed the presence of a post-fire peak in both shrub cover strata approximately seven years post-fire, which rapidly senesced to mean levels. Secondly, canopy cover from trees greater than 30 m was unaffected by most fires. This study contributes to a large body of work into post-fire vegetation regrowth modelling using both remote sensing and lays the groundwork for further study using in situ sampling methods, and fuel load and fire behaviour models.

FoR codes (2020)

410205 Fire ecology



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.