Ignition and burning mechanisms of live spruce needles

Highlights

• Dead vs live forest fuels burning behavior was investigated.

• IR-Schlieren hybrid was used to visualize the burning live-fuels jetting.

• Scaling using the law approach was conducted to formulate governing Pi numbers.

• Only live fuels exhibit jet-flame behavior that assists in sustaining the flame.

• These jets result in micro-spotting, ejecting ignited fuel out of the burning zone.

Abstract

Live foliage for some tree and shrub species can support flaming fire spread at much higher moisture content than dead fuel materials. However, the role of live fuels in forest fires has been controversial in the past decades. Although ignition and spread statistical data for live and dead fuels exist in the literature, a clear understanding of the fundamental difference in the burning behavior is missing. To illuminate the role of live fuel on forest fire spreading, a laboratory ignition experiment was designed to examine the burning behavior of live Norway spruce needles. A Schlieren-Infrared combined measurement apparatus was developed with a spatial resolution of 0.75 mm and a time resolution of 0.0025 s, to visualize/measure the ignition behavior of live fuels. Schlieren and IR images revealed that the ejection of live fuel volatiles can alter the flame direction and induce previously unaccounted heating of the nearby fuel. Depending on the conditions, these interferences could heat and modify the heat flux received by the adjacent fuels. To analyze each of these outcomes, a scaling analysis using the law approach was performed. First, theoretical equations were developed and validated against a set of previously published experimental data. After the characteristic equations were verified, we used them to assess the volatile ejection phenomenon. We found that adjacent fuels were preheated by hot volatiles ejected from the heated live needle, and direct flame contact ignited the adjacent fuels. Our IR experiments confirmed the outcomes of the scaling analysis. The rapid ejection of volatiles was also found to propel burning needles far from the burning branch, resulting in micro-spotting.

Introduction

Wildland fires have become increasingly destructive, and the associated damage and firefighting costs have increased over the past years. For example, in the U.S. alone, the annual fire suppression budget went from $580 million for the period (1991–2000) to $1.2 billion annually from 2001 to 2010, motivating more fire research [1]. Despite the increased efforts and budget to fight wildland fires, the severity and frequency of wildland fires increased over the past decades [2], [3]. An accurate prediction of fire spread rate through wildland fuel beds is important but challenging due to the many different variables involved. Still, firefighters need to predict fire spreading direction and spread rate for their practical firefighting strategy and practice even with the associated uncertainty and fire management personnel to perform strategic planning and decision making [4].

Conifer trees are known to withstand extreme and harsh environmental conditions and are found in a wide range of habitats around the world [5]. Conifer forests extend across continents in the Northern hemisphere [6], [7], [8], [9]. This makes the study of conifer fires a worldwide interest, as the ability to protect these natural landscapes is a matter of interest to many countries across the globe. Conifer forests are also susceptible to burning as crown fires, which are characterized by their high intensity and fast rate of spread. Crown fires pose high risks to human populations and natural resources due to a large amount of heat-release and high spreading rates, but the fire spreading mechanism of crown fires is poorly understood [10] due partly to the unknown combustion mechanisms of live crown fuels.

Historically, researchers assumed that live fuels behave as wet dead ones [11]. Studies focused on the moisture content in foliage and its effect on fire spread with that assumption in mind. For instance, Rothermel [12] modeled fire spread in wildland fires, including the effect of fuel moisture content without distinguishing the live and dead fuels. Since then, many researchers have investigated the effect of live fuel moisture content on fire ignition and spread. For instance, in addition to Sun et al. [13], who linked live fuel moisture content (LFMC) with the ignition and spread features of fire, Dennison and Moritz [14] studied the impact of seasonal changes in moisture content over the years and their influences on the fire ignition. More recently, Alexander and Cruz [15] suggested that the normal range of live fuel moisture content (LFMC) has little effect on fire spread rate. Nevertheless, researchers like Rossa and Fernandes [16] examined the monthly values for dead and live fuel moisture content and correlated them with each other, showing why the effect of LFMC on fire rate of spread has been vague, although it does have an influence. Furthermore, Anderson et al. [17] claimed that the correlation between LFMC and the height and bulk density might be the reason for the low effect of LFMC. Although the effect on LFMC was unclear and controversial among researchers, a distinction between live and dead fuels was early demonstrated by Hawley [18] and then followed by Weise et al. [19], where the former’s study showed that dead fuel would not sustain flame at moisture contents higher than 35%, and the latter found that fire could spread in live fuels even with doubled or tripled moistures. One reason might be that minor chemical components existing in live spruce needles (terpenes), as seen in [20], have an energy content of approximately 45 MJ/kg, higher than that of gasoline and diesel [21]. However, other studies noted the burning of live fuel leaves with popping and snapping sounds [10]. Even a small burst in the surface of the foliage was noted in some of the experiments conducted by Fletcher et al. [22]; this was tied to the ability of live fuel casing to retain the moisture content inside until its structure is ruptured and the evaporated moisture is violently released [11]. However, the effect of this burning behavior on the fire spread problem has not been studied. The above-mentioned literature suggests that our current understanding of the difference between dead and live fuels is incomplete and Pimont et al. [23] suggests there is an urgent need for understanding the effect of LMFC on fire spread behavior.

There is a rich literature that statistically studied the differences between live and dead fuels in terms of the fire spread rate and ignition. However, a basic understanding of the fundamental differences between the burning behaviors exhibited by the two is still missing [16], [19], [24]. To answer that question, we developed a hybrid schlieren-IR visualization setup that visualizes the ignition behavior of a freshly cut Norway spruce branch vs. a geometrically similar dead fuel.

The phenomena of “micro-explosions,” previously reported as popping sounds, were visualized in live but not dead fuels as expected. These micro-explosions occur as the fire approaches, due to the pressure build-up inside the needle when its moisture evaporates until the needle eventually yields releasing jets of hot volatiles. Characterized by a mixture of water vapor and flammable monoterpenes, those jets can sustain and expand the flame, as visualized by the IR thermography. The following three points summarize the contributions of this paper to the progress in the fire spreading through forest fuel beds.

• A combined Schlieren-IR visualization of the burning behavior of live vs. dead fuels was performed to explain previously reported live fuel mysteries including the newly observed effect of “micro-explosions” in live fuel.

• Scaling and IR analyses were conducted based on the above visualization results to theoretically assess the effect of micro-explosions on preheating unburned adjacent spruce needles.

• A newly observed “micro-spotting” induced by the micro-explosions ejecting ignited needles outside the burning zone comprising new burning zones.

The next section will discuss the details of the experimental methods, sample preparation, accuracy, and reproducibility.