Estimation of Fire Danger in Hawai‘i Using Limited Weather Data and Simulation.

Date
2010-04
Authors
Weise, David R.
Stephens, Scott L.
Fujioka, Francis M.
Moody, Tadashi J.
Benoit, John
Contributor
Advisor
Department
Instructor
Depositor
Speaker
Researcher
Consultant
Interviewer
Journal Title
Journal ISSN
Volume Title
Publisher
Honolulu, University of Hawaii Press
Volume
Number/Issue
Starting Page
Ending Page
Alternative Title
Abstract
The presence of fire in Hawai‘i has increased with introduction of nonnative grasses. Fire danger estimation using the National Fire Danger Rating System (NFDRS) typically requires 5 to 10 yr of data to determine percentile weather values and fire activity. The U.S. Army Po¯ hakuloa Training Area in Hawai‘i is located in the interface zone between windward and leeward weather conditions and needed to develop fire danger values but did not have sufficient weather or fire occurrence data. Use of simulation to estimate fire danger (expressed as fire risk) for areas with limited weather data was investigated. Influence of spatial resolution of weather information on fire risk was examined by comparing fire risk calculated using one or three weather stations and gridded weather predictions from the Mesoscale Spectral Model. Predicted gridded temperature was positively correlated with observed temperature; predicted and observed relative humidity were not significantly correlated. Simulated fire risk differed between weather data percentiles and between weather data resolutions. Predicted risk estimated from gridded weather data agreed more closely with observed risk estimated from weather data observed at all three remote automated weather stations. Correlation between simulated fire risk and the NFDRS Ignition Component was statistically significant for the single weather station simulations. Correlations between risk and the Ignition Component were not statistically significant for the three station and gridded weather data scenarios, which illustrates the difference between fire danger determined at broad spatial scales and fire risk resolved at finer spatial scales. Fire spread simulation modeling to estimate fire risk in areas with limited historical weather and fire occurrence data can provide finer-scale information than the NFDRS, which is better suited to larger, homogeneous areas with more complete fire and weather data. Values for the NFDRS Burning Index were determined and incorporated into the wildland fire management plan for Po¯ hakuloa Training Area.
Description
v. ill. 23 cm.
Quarterly
Keywords
Citation
Weise DR, Stephens SL, Fujioka FM, Moody TJ, Benoit J. Estimation of Fire Danger in Hawai‘i Using Limited Weather Data and Simulation. Pac Sci 64(2): 199-220.
Extent
23 p.
Format
Geographic Location
Time Period
Related To
Rights
Rights Holder
Email libraryada-l@lists.hawaii.edu if you need this content in ADA-compliant format.