SAFETYLIT WEEKLY UPDATE

We compile citations and summaries of about 400 new articles every week.
RSS Feed

HELP: Tutorials | FAQ
CONTACT US: Contact info

Search Results

Journal Article

Citation

Troy A, Moghaddas J, Schmidt D, Romsos JS, Sapsis DB, Brewer W, Moody T, Troy A, Moghaddas J, Schmidt D, Romsos JS, Sapsis DB, Brewer W, Moody T. Int. J. Wildland Fire 2022; 31(6): 586-598.

Copyright

(Copyright © 2022, International Association of Wildland Fire, Fire Research Institute, Publisher CSIRO Publishing)

DOI

10.1071/WF21176

PMID

unavailable

Abstract

Despite the intensity of the 2018 Camp Fire, many structures survived in heavily burned areas. Logistic regressions were run to determine which structural and parcel characteristics predicted structure survival using two data sets. The first, CAL FIRE's Damage Inspections (DINS) dataset, included 14 518 destroyed and 622 partially damaged structures. The second, combining information from the DINS and Defensible Space (DINS+DSPACE) databases, had many more attributes and was better balanced between destroyed (n = 728) and surviving (n = 676) structures, but was much smaller. Several approaches were compared for filtering out records with null values.

RESULTS were largely consistent with previously literature, finding that structural hardness factors (e.g. double-paned windows, enclosed eaves, ignition-resistant roofs and siding, no vents, etc.) are important in determining structure survival. Newer structures, built after California's recent (2005 and 2007) fire safe building code updates, were more likely to survive, as were homes with higher improvement values. Mobile homes were far more likely to be destroyed. The role of fuel mitigation around structures was less conclusive; defensible space clearance had only a weak association with structure survival, although DINS+DSPACE results suggested a slight reduction in risk due to removing leaves and needles from gutters/roofs and keeping surrounding dead grass mowed.


Language: en

NEW SEARCH


All SafetyLit records are available for automatic download to Zotero & Mendeley
Print