Experimental and Theoretical Investigation of Fuel Bed Ignition by Embers and Heated/Burning Particles

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

1066520 Fernandez-Pello, Carlos According to US Fire Administration statistics 96,000 wildfires burned 9.9 million acres in 2006, greater than the entire area of Maryland and the DC metro area. In 2007, "outdoor and other" (primarily wildland) fires resulted in 45 deaths, 650 injuries, and $2.6 billion in property damage. Many wildland and wildland urban interface (WUI) fires are ignited by hot metal particles or embers generated by powerline interactions, hot work/welding, overheated catalytic converters, seized train brakes, and other sources of incandescent particles. Firebrand (ember) spotting is the primary vector for spread of wildland and WUI fires under dry, hot, and windy conditions that produce the most devastating fires. Spotting leads to rapid fire spread because embers are lofted by fire plumes and transported downwind to ignite secondary fires or structures remote from the flame front. Most structures destroyed during WUI fires are ignited by firebrands that penetrate vents or ignite roof construction, not direct flame impingement. Civilians and firefighters alike can become trapped between spot fires with no escape route; partly to blame is the difficulty associated with accurate prediction of spotting and its effect on fire spread. The conditions under which embers and heated particles can ignite a spot fire are not well understood. The work proposed here aims to develop the science needed to accurately predict the conditions leading to ignition of fuel beds by embers and heated particles. It is a combined experimental, analytical, and numerical modeling study of ignition of fuel beds by embers and heated particles. A small-scale wind tunnel will be used to experimentally investigate the ignition of both "laboratory" and "real world" fuel beds by woody embers/firebrands and heated, molten, or burning metal particles. Experiments will be used to predict ignition probability as a function of fuel bed and ember/heated particle properties and a physics-based numerical model of fuel bed ignition by embers and heated particles, will be developed. The information will then be incorporated in predictive models of wildland fire propagation. Intellectual merit: As a physical process, ignition of fuel beds by an ember or hot metal particle is one of the richest problems in fire and combustion science, incorporating solid and gas phase thermosciences processes. This project will unravel the controlling physics and chemistry, thereby developing significant new knowledge. The proposed work is transformational because it will, for the first time, provide a quantitative understanding of ignition of fuel beds by embers and hot particles grounded in rigorous experiment, theory, and numerical modeling. The experimental program will a detailed and well-controlled study into ignition of fuel beds by embers and heated particles conducted to date. Broader impacts: This work has practical applications that will translate into real-world economic and life safety benefits. The potential benefit to society of this work stems from its ability to reduce devastating fire losses when incorporated into predictive models and disaster planning as well as its potential to be used in protocols for preventative fuels treatments, development of smarter regulatory guidance and test standards for fire-safe construction, staging orderly evacuations, and efficient allocation of fire suppression resources during fire events. By reaching out to organizations that promulgate codes and standards such as the California Public Utilities Commission, ASTM E5, the International Code Council, and CalFire, it will be ensured that this research makes a real world impact. This research will be incorporated into the Combustion Processes course at UC Berkeley and undergraduate students will be invited to participate in the research. On average, four undergraduate students conduct research in Prof. Pello?s laboratory often from programs aimed to attract underrepresented minorities to engineering graduate education, such as NSF-LAMP. The results of this research will be broadly disseminated to enhance scientific and technological understanding through presentations at national and international conferences, conventional publication (peer-reviewed archival journal articles), and "Science 2.0" methods that involve electronically disseminating papers, reports, guides, data, simulation results, experimental data, photographs/videos, source code, executable files, and model output produced as part of this grant.