Please use this identifier to cite or link to this item:
Modeling fuels and wildfire behavior in Hawaiian ecosystems
|dc.contributor.advisor||Litton, Creighton M.|
|dc.subject||Natural resource management|
|dc.title||Modeling fuels and wildfire behavior in Hawaiian ecosystems|
|dc.contributor.department||Natural Resources and Environmental Management|
|dcterms.abstract||The interactions of nonnative invasions and landscape disturbances on novel fuel beds exacerbate the impacts of wildfire in Hawaii, and throughout the tropical Pacific Islands. Intensive field sampling and hybrid remote sensing techniques are valuable means of assessing regional and local variability in fuels in response to environmental gradients and management actions. The first chapter of this thesis introduces the context of these challenges, and provides a brief study overview. The second chapter seeks to understand the effects of nonnative feral ungulate removal, as well as the effects of ecological restoration, on fuels and modeled wildfire behavior in Hawaiian terrestrial ecosystems along a precipitation gradient. I sampled fuel characteristics inside and outside of 13 ungulate exclosures across a 2740 mm mean annual precipitation (MAP) gradient, and used linear mixed effects modeling to identify environmental and management drivers of fuel dynamics and modeled wildfire behavior. Fine fuel loads increased after ungulate exclusion (up to 11.3 Mg ha-1), and increased in magnitude with MAP. Fuel load differences resulted in higher intensity modeled wildfire behavior (up to 1.9 m difference in flame length), which also increased in magnitude with MAP. Ecological restoration of sites following ungulate removal, however, decreased fine fuel loads over time by as much as 41% after ten years. Results from the second chapter demonstrate a clear trend in increased fuel loads after ungulate removal across a wide precipitation gradient, and point towards the need for effective fuels management strategies, such as long-term active ecological restoration, to reduce fuel loads and invasive species cover so as to break the positive feedback between nonnative invasions and wildfires. The third chapter of this thesis evaluates the accuracy of landscape fuel mapping by the national geospatial fuels product LANDFIRE versus a custom mixed methods approach that utilized supervised classification of field sampled fuel data, biophysical predictors, and remote sensing over a heterogeneous, dry tropical montane landscape in Hawaii. The custom fuel map demonstrated only a 27% agreement with the standard LANDFIRE fuel map, but had a 58% overall mapping accuracy which compares favorably with mapping accuracies from other fuel studies. In line with previous research in temperate ecosystems, my analysis indicates LANDFIRE can provide a first approximation of fuel conditions and wildfire risk for understudied regions, but that supervised classification of local fuels data, biophysical predictors, and remote sensing can greatly improve accuracy and utility. In the final chapter, I offer suggestions for negotiating tradeoffs between conservation, restoration, and limited wildfire suppression resources. Results from both data chapters demonstrate that fuel loads in Hawaii are highly variable, even within similar vegetation types, and substantially driven by climate, land use, disturbance history, and management actions. Limiting wildfire risk in the Hawaiian Islands requires effective fuel reduction treatments after nonnative feral ungulate removal, particularly in mesic and wet areas, as well as continued calibration of fuel maps.|
|dcterms.publisher||University of Hawai'i at Manoa|
|Appears in Collections:||
M.S. - Natural Resources and Environmental Managament|
Please email email@example.com if you need this content in ADA-compliant format.
Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.