Christmas Tree Fire Experiments at the University of Maryland

Introduction

Each year since 2014, the Department of Fire Protection Engineering at the University of Maryland (UMD) has coordinated a Christmas tree fire safety event to highlight the potential fire risk of dry Christmas trees. Full-scale fire experiments at these events are used to show how properly maintaining a cut Christmas tree (to maintain a high moisture content in tree branches and needles) can help to limit ignition likelihood, fire growth rate, and peak fire size. The event is regularly covered by local and national media (see Fig. 1) and the experiments also provide a valuable learning experience for students of the department.

In its early years, this fire safety event served as the final for an introductory fire science program taught at UMD (this program was supported in 2015 by a grant from the Society of Fire Protection Engineers, SFPE, Scientific and Educational Foundation). More recently, senior- and graduate-level students of the department have been invited to experience these fire experiments from a safe distance to contextualize the fire behaviors that they study. In various years, these experiments were included in courses such as fire dynamics, enclosure fire modeling, and the graduate laboratory to allow students to practice assessing design fires, calculating fire risk (e.g., zone of influence and heat transfer at a distance), and performing statistical and uncertainty analyses of large fire datasets.

Additional tests have been performed as part of this fire safety event to explore the impact of residential sprinklers on tree burning behavior, to demonstrate bench-scale burning of natural versus artificial tree branches, and to quantify factors affecting firebrand generation (known to strongly impact wildfire spread (Manzello et al., 2020). In 2019, as part of a collaborative effort between researchers at the National Institute of Standards of Technology (NIST), a comprehensive test series was conducted to characterize firebrand generation during the burning of full-scale vegetation in the absence of ambient wind. Measurement data from the 2019 experiments include: time-resolved mass loss rate during burning, heat flux at a distance, and firebrand yield (i.e., mass of firebrands generated per mass of tree burned). Further analysis of the more than 10,000 firebrands collected during these experiments provided insight regarding the size, mass, and shape distributions of these firebrands and how these firebrand characteristics and measured firebrand yields depend on tree moisture content and species (Leventon et al., 2024).

The Competition

The data collected during these experiments are also used to enable a friendly competition that encourages communication and collaboration between members of the fire safety science community. Beginning in 2014 as a small event within the Department of Fire Protection Engineering at UMD, the 2023 Competition (tenth anniversary) received 205 predictions from 72 unique institutions across 25 countries (and 11 US states); a world map of countries and US states that participated in 2023 is provided in Fig. 2. 

Figure 3 highlights results throughout the first ten years of the competition.  As seen, the number of submissions has grown each year. Substantial changes – shape/variety of time-resolved heat release rate (HRR) curves and total number of predictions – occurred in 2017 when a web-based HRR curve generator was first introduced. This custom-made app allows participants to ‘build’ (and submit) their own fire by defining just a few parameters (e.g., Peak Heat Release Rate [kW], Relative time to Peak HRR, and Total Heat Release [MJ]). The HRR curve generator has been updated for the 2024 competition to provide more representative HRR profiles and to simplify the submission process.

The 2024 Competition

This year’s event (2024) is unique. In collaboration with NIST, the competition burns will consist of repeated experiments on extremely large (5.8 m tall) Douglas-Fir trees (see Fig. 4) burned in the National Fire Research Laboratory (NFRL). Measurement data, video recordings, and a full test report will be released the day of the competition. These experiments are part of a larger test series of design fires to be included in the NIST Fire Calorimetry Database.

For this competition, students, faculty, researchers, and engineers from fire safety programs around the world are invited to predict the burning behavior – i.e., time-resolved fire size (heat release rate, HRR) – of a dry Christmas tree. The highest individual and highest group scores are announced to all participants, and the two research teams (3 or more individual predictions from the same institution) that earn the highest team scores earn the coveted Golden and Silver Pinecones (see Fig. 5).

Although this competition is meant to be a fun event for the community, in addition to sharing an important holiday fire safety message, it also offers a teachable moment: experimental measurements have an inherent uncertainty and our ability to accurately predict fire behavior should always be assessed with explicit considerations for this uncertainty. Thus, predictions are scored with respect to the average and standard deviations of repeated measurements (repeatability), and they also explicitly account for measurement device uncertainty. Participants earn credit for accurately predicting key aspects of burning behavior (i.e., peak HRR, time to peak HRR, overall HRR profile, and total energy release).

2024 Competition Details

This year’s competition will take place on Thursday, December 19, 2024, at 12:00 pm (EST). All HRR predictions received prior to this date will be scored for the competition.

The submission (and generation of) predicted HRR curves is made possible by visiting https://pages.nist.gov/christmas_tree_fire_safety/. On this page, participants can ‘build’ (and submit) their own Christmas tree fire HRR predictions for this year’s competition by adjusting three parameters that define the HRR: Peak heat release rate (Peak HRR, kW), relative time to Peak HRR, and total heat released (MJ). Further details of how to use this curve generator and submit predictions are available online.

Video recordings of these experiments will be shared during a community watch party/livestream at noon on the day of the competition: https://umd.zoom.us/j/7239055335. The final details regarding how to watch the event will be provided to all participants who submit a prediction.

Further competition details are available online: https://fpe.umd.edu/burn-competition.

Fire Safety Statistics

According to National Fire Protection Association (NFPA) statistics, between 2016 and 2020, fire departments in the United States responded to an average of 160 home structure fires that began with Christmas trees (Campbell, 2022). These fires caused an annual average of two civilian deaths, 11 civilian injuries, and $12 million in direct property damage. Fortunately, current statistics indicate that the estimated number of Christmas tree fires has significantly decreased (from a high of 850 in 1980, to 270 in 1998, to 160 more recently).

Although these events represent a small fraction of fires reported each year (Campbell, 2022) – less than 0.1% of all fires reported in the National Fire Incident Reporting System, NFIRS (partly because these fires are seasonal, as Christmas trees are typically found in homes only for a short time each year) – these fires are particularly challenging to life safety. Specifically, they have an outsized impact on direct property damage (Campbell, 2022) and, based on data collected between 2011 and 2015 (Ahrens, 2017), when Christmas trees were the first item ignited in home structure fires, one out of every 32 reported fire events resulted in a death, compared to an average of one death per 143 total reported home fires (this represents a factor of 4.4x increase in mortality).

In 2020, the American Christmas Tree Association (ACTA) reported that nearly 94 million households had at least one Christmas tree, of which 15 percent were natural (vs. artificial trees) (ACTA, 2020). However, a recent NFPA report (Ahrens and Maheshwari, 2020) suggests that fires involving natural trees are nonetheless approximately twice as common as those involving artificial trees. Many of these fire events occur later in the season (Campbell, 2022; Ahrens, 2017; NFPA 2013), potentially because the longer a natural tree is kept, the more likely it is to dry out, which can increase the potential for ignition, faster fire growth, and larger peak fire size (Ahrens, 2017: ATF, 2015; Stroup et al., 1999; Hoehler et al., 2020)  (see Video 1 at the very beginning).

Christmas Tree Fire Safety Tips

If you plan on keeping a natural Christmas tree this season, please keep it healthy, well-watered, and away from potential ignition sources. The National Fire Protection Association (NFPA) suggests several additional steps that you can take to reduce the risk of a Christmas tree fire in your home (NFPA 2020):

• Choose a tree with fresh, green needles that do not fall off when touched.

• Before placing a tree in its stand, cut 5 cm (2 in.) from the base of the trunk;

• Add water to the tree stand. Be sure to add water daily.

• Make sure that the tree is at least three feet away from any heat source (e.g. space heaters, candles, fireplaces, heat vents, or lights).

• Make sure that the tree does not block an exit.

• Use lights that are listed by a qualified testing laboratory. Replace any string of lights with worn or broken cords or loose bulb connections.

• Always turn off tree lights before leaving home or going to bed.

• Never use lit candles to decorate the tree.

• Get rid of the tree after Christmas or when it is dry and keep it away from your home/garage.

• Check with your local community to find a recycling program.

NFPA also provides other fire safety tips for the holiday season:
https://www.nfpa.org/education-and-research/home-fire-safety/winter-holidays

Join the contest and make it the most competitive Christmas Tree Burn ever!

Isaac Leventon

References

Ahrens, M., “Home Structure Fires Involving Christmas Trees,” NFPA Research. (2017)

Ahrens, M., Maheshwari, R., “Christmas Tree Fires,” NFPA Research. (2020)

American Christmas Tree Association (ACTA), “Ninety-four Million U.S. Households Will Celebrate The Holidays With At Least One Christmas Tree,” from November 2020 Nielsen English Language Panel Views Survey. Accessed September 27, 2021.

Campbell, R., “Christmas Tree Fires,” NFPA Research. (2022)

Bureau of Alcohol, Tobacco, Firearms and Explosives, “Executive Summary: Fire at 936 Childs Point Road”, ATF Investigation #761010-15-0026. (2015)

Leventon, I.T., Tlemsani, M., Hajilou, M., Ju, X., Gollner, M. “Generation of Firebrands from the Burning of Full-Scale Vegetation in the Absence of Ambient Wind”, International Association of Fire Safety Science 2024.

Hoehler, M. S., Bundy, M. F., DeLauter, L. A., Gerskovic, L., Garcia, J. R., “Fire Hazards of Dry Versus Watered Christmas Trees,” NIST Technical Note 2131. (2020)

Manzello, S. L., Suzuki, S., Gollner, M. J., & Fernandez-Pello, A. C. “Role of firebrand combustion in large outdoor fire spread”. Progress in energy and combustion science, 76. (2020)

NFPA, “Christmas Tree Safety” Available: https://www.nfpa.org/-/media/Files/Public-Education/Resources/Safety-tip-sheets/ChristmasTreeSafetyTips.ashx (2020). Accessed November 22, 2024.

NFPA Fire Analysis & Research Division, “NFPA Christmas Tree Fact Sheet,”. http://www.nfpa.org/news-and-research/fire-statistics-and-reports/fire-statistics/fire-causes/holiday/christmas-tree-and-holiday-lights. (2013). Downloaded December 7, 2014.

Stroup, D. W., Delauter, L., Lee, J., Roadarmel, G., “Scotch Pine Christmas Tree  Fire Tests - Report of Test 4010,” Available: https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=907753  (1999)