Document


Title

Using a prescribed fire to test custom and standard fuel models for fire behaviour prediction in a non-native, grass-invaded tropical dry shrubland
Document Type: Journal
Author(s): Andrew D. Pierce ; Sierra McDaniel ; Mark Wasser ; Alison Ainsworth ; Creighton M. Litton ; Christian P. Giardina ; Susan X. Cordell
Publication Year: 2014

Cataloging Information

Keyword(s):
BehavePlus; fire behaviour; fire management; flame length; flame length; fuel management; fuel model; fuel models; fuel moisture; grasses; Hawai'i Volcanoes National Park; invasive grasses; invasive species; national parks; range management; rate of spread; rate of spread; shrublands; tropical regions
Region(s):
Record Maintained By:
Record Last Modified: October 25, 2018
FRAMES Record Number: 53127
Tall Timbers Record Number: 30440
TTRS Location Status: Not in file
TTRS Call Number: Available
TTRS Abstract Status: Fair use, Okay, Reproduced by permission

This bibliographic record was either created or modified by the Tall Timbers Research Station and Land Conservancy and is provided without charge to promote research and education in Fire Ecology. The E.V. Komarek Fire Ecology Database is the intellectual property of the Tall Timbers Research Station and Land Conservancy.

Description

Questions: Do fuel models developed for North American fuel types accurately represent fuel beds found in non-native, grass-invaded tropical dry shrublands? Do standard or custom fuel models used in fire behaviour models with in situ or remote automated weather stations (RAWS) measured fuel moistures affect the accuracy of predicted fire behaviour in grass-invaded tropical shrublands? Location: Hawai'i Volcanoes National Park, Hawai'i, USA. Methods: Pre-fire fuel loads of coarse woody debris, live herbaceous and live woody fuel loads were quantified with Brown's transects and biomass sampling to create a custom fuel model for non-native grass-invaded tropical dry shrublands in Hawai'i. In situ fuel moistures were quantified using oven-dried vegetation samples, and compared to RAWS estimated fuel moistures. Fire behaviour was recorded on a stationary video camera to quantify flame length (FL) and rate of spread (ROS). Observed fire behaviour was compared to BehavePlus predicted fire behaviour parameterized with both standard and customized fuel models, and in situ and RAWS-based estimates of fuel moisture. Results: The custom fuel model and measured fuel moistures performed better than most standard models, but over-predicted actual ROS and top decile FL by 29% and 26%, respectively. The best match between observed and modelled fire behaviour came from a standard fuel model for shrublands with a grassy matrix (23% under-prediction for ROS and 9% under-prediction for FL) using measured fuel moistures. Using fuel moistures and wind speeds estimated from the nearest RAWS station (5 km from the fire) substantially decreased prediction accuracy of the custom fuel model and increased its relative error to 71% over-prediction of ROS and 45% over-prediction of FL. Conclusions: Fire behaviour in at least some tropical fuel beds can be accurately modelled using certain standard or custom fuel models. Standard fuel models should not be applied uncritically to systems outside of North America, as our comparison showed widely ranging accuracy across six standard models. In addition, the current reliance on RAWS data for meteorological inputs to predict fire behaviour in the tropics, especially in the US-affiliated tropical Pacific, must be used with caution. Instead, field-measured fuel moistures should be used when possible. © 2014 International Association for Vegetation Science.

Citation:
Pierce, A. D., S. McDaniel, M. Wasser, A. Ainsworth, C. M. Litton, C. P. Giardina, and S. Cordell. 2014. Using a prescribed fire to test custom and standard fuel models for fire behaviour prediction in a non-native, grass-invaded tropical dry shrubland. Applied Vegetation Science, v. 17, no. 4, p. 700-710. 10.1111/avsc.12111.