U.S. DEPARTMENT OF COMMERCE
NATIONAL INSTITUTE OF STANDARDS AND TECHNOLOGY
Gaithersburg, MD 20899

REPORT OF TEST
FR 4010
December 1, 1999
Scotch Pine Christmas Tree Fire Tests
D.W. Stroup, L. DeLauter, J. Lee and G. Roadarmel
Building and Fire Research Laboratory
National Institute of Standards and Technology
U.S. Department of Commerce
Gaithersburg, MD 20899

A series of fire tests were conducted to characterize the potential hazard from ignition of Scotch
Pine Christmas trees. Heat release rate was measured as a function of time from ignition using a
large calorimeter. In addition, radiative flux and total heat flux were measured at a location 2 m
from the tree centerline. Seven of the trees were allowed to dry for approximately three weeks.
These trees burned easily when ignited with an electric match. An eighth tree, freshly cut and
kept in water until just before testing, could not be ignited.
Key Words:
fire data; fire models; fire tests; heat release rate; radiation heat flux

Measurement of the rate at which a burning item releases heat is a critical parameter in fire
protection engineering. The heat release rate can be used in the characterization of the hazard
represented by a given fuel package. Heat release rate can provide information on fire size and
fire growth rate. When used as input to a computer fire model, the heat release rate can be used
to estimate available egress time and determine suppression system activation time. Heat flux
measurements can be used to estimate potential for ignition of adjacent fuel items.

Christmas trees have been involved in a number of significant fire incidents [1, 2, 3, 4, 5].
Typical ignition scenarios involve shorted electrical lights [1, 2, 3] or open flame exposure [4].
According to the U.S. Consumer Product Safety Commission, Christmas trees account for
approximately 400 fires annually, resulting in 10 deaths, 80 injuries and more than $15 million in
property damage [6].

There are two primary types of trees used every year as Christmas trees; they are either real or
artificial. Real trees vary significantly in species, shape and size. The moisture content of these
trees and the dryness of their needles can have a significant impact on ignition and heat release
rate. While an effort was made to obtain reasonably similar trees, the trees used in this study did
vary somewhat. In order to address this variation, a number of replicate tests were conducted
during this study. Differences in the heat release rate and heat flux data can also result from
random differences in the spread of the fire. This study focuses primarily on the heat release rate
of relatively dry trees. As a point of comparison, one tree used in this study was maintained in
accordance with local regulations governing use of real trees in offices [7]. Among other things,
these regulations require that the tree be cut in the user's presence, have an additional two inches
cut from the bottom of the trunk, and be placed in a stand with at least a 7.6 liter capacity. In
addition, the user must maintain the tree's water on a daily basis and keep a record of the
maintenance.

The experiments were conducted under the main hood in the NIST Large Fire Research Facility.
This hood is 4 m by 5 m and slopes upward to a 1.2 m square duct. During a fire test, data from
various sensors is acquired using a computer-based data acquisition system. The fire test data is
recorded on magnetic media for further data reduction and interpretation after the test. Data
acquisition and reduction in the Large Fire Research Facility are accomplished using in-house
developed computer software [8].

Using the principle of oxygen consumption, it is possible to calculate the heat release rate of
burning materials when the products of combustion are collected in an exhaust hood. Parker [9,
10] presents several sets of equations for calculating heat release rate using oxygen consumption.
The appropriateness of each set of equations depends on the combustion products being
measured. A paper by Janssens [11] proposes a form of the equations for calculating heat release
rate specifically for full-scale fire test applications.

Heat release rate is determined in the NIST Large Fire Research Facility using the equations
from reference 10 together with data obtained from instruments in the exhaust hood. The
measured heat release rate has been shown to be within 20 percent of the actual value [12].
Reference 12, page 8 contains details concerning the calculation of heat release rate and its
implementation in the Large Fire Research Facility.

Each experiment was conducted with the tree centered under the exhaust hood (Figure 1). The
radiative heat flux and total heat flux were measured with water cooled Gardon type transducers.
The total heat flux gauge was located 1.2 m above the floor with its face 2 m from the tree
centerline. The face of the radiative heat flux transducer was also located 2 m from the
centerline of the tree. However, it was located 50 mm vertically above the total heat flux
transducer. Based on manufacturer's data, the standard uncertainty for the heat flux
measurements is estimated at ±3 % [13].

Eight Scotch pine tree fire experiments were conducted under the main hood in the Large Fire
Research Facility. With the exception of the last tree, all of the trees had been allowed to dry for
approximately 3 weeks in a room with a nominal air temperature of 23 °C and a relative
humidity of 50 %. The last tree was conditioned in the same room, however, it was placed in a
19 liter bucket filled with water. Observation of the water level in the bucket suggested that the
tree continued to absorb water until the test. Prior to testing, the moisture content in each tree
was measured at several points on the trunk using a moisture meter. The moisture content
ranged from 25 % to 35 %. While the moisture content of the tree trunks varied by only 10
percent, the first seven trees differed significantly in apparent dryness from the eighth tree. The
needles on the first seven trees were dry needles to the touch and slightly brown in color. The
needles broke easily when bent slightly and fell from the tree when the branches were shaken.

An eighth tree was maintained in accordance with local regulations governing use of real trees in
offices [7]. Among other things, these regulations require that the tree be cut in the user's
presence, have an additional two inches cut from the bottom of the trunk, and be placed in a
stand with at least a 7.6 liter capacity. In addition, the user must maintain the tree's water on a
daily basis and keep a record of the maintenance. The eighth tree had a much greener color and
the needles were very pliable. When a needle was pulled, the tree would bend slightly before
needle would separate from the branch.

Each tree was weighed and measured before the test. In addition, the weigh of the remainder of
the tree was determined after the test. This data is presented in Table 1. The standard
uncertainty associated with the weigh measurements is ±0.2 kg, and the uncertainty associated
with the length measurements is ±6 mm.

* The width is measured at the widest point of the tree.

Each tree was attached at the bottom of the trunk to a 0.5 m square by 12.7 mm thick plywood
board using a 76 mm drywall screw. This board provided the necessary support to keep each tree
standing during the tests. An electric match placed in the lower branches of the tree was used as
the ignition source for each test. An electric match consists of a typical paper match from a
matchbook wrapped with 0.3 mm diameter nichrome alloy wire. When an electric current passes
through the wire, the wire generates heat due to resistance heating. This heating causes ignition
of the match. If the tree is dry enough, the matches should ignite the tree. An electric match and
its placement in a tree are shown in Figures 2 and 3 respectively.

A single match was insufficient to produce ignition of the eighth tree. Ignition was attempted a
second time using an electric match composed of an entire matchbook containing 20 paper
matches (Figure 2). This second attempt also failed to produce ignition. Finally, an open flame
was applied to the tree using a propane torch. The branches ignited briefly, but they selfextinguished
upon removal of the torch from the branches.

The heat release rate curves obtained as a function of time from ignition for each of the first
seven fire tests are shown in Figures 4 - 10. Since the eighth tree could not be ignited, there is no
heat release rate curve available for the eighth test. Peak heat release rates range from about 1.6
MW to 5.2 MW. The total heat flux (solid line) and radiative heat flux (dashed line)
measurements for each of the first seven tests are shown in Figures 11 - 17. Again, there is no
heat flux measurements for the eighth test since the tree did not ignite. Photographs of the trees
prior to testing and at the point of peak heat release rate for the second through seventh tests are
shown in Figures 18 - 29. The arrows in the figures indicate the ignition location for each test.
Figure 30 depicts the eighth tree prior to testing, and Figure 31 illustrates the results of several
attempted ignitions. Due to an equipment failure, photographs of the first tree and test are not
available.

[1] Borg, L. B., “Christmas Tree Fire Scars Utah's Historic Governor's Mansion,” State Fire
Marshal, UT, Sprinkler Age, Vol. 14, No. 7, 10,12, July 1995.

[2] Lathrop, J. K., “Fire Destroys Yacht Club, Kills Three - Natural Christmas Tree Erupts in
Flames,” Fire Journal, Vol. 71, No. 5, pp. 21-23, September 1977.

[3] David, J., “Seven-Fatality Christmas Tree Fire, Canton, Michigan, December 22, 1990,”
USFA Fire Investigation Technical Report Series Report 046, Federal Emergency
Management Agency, Washington, DC, 1991.

[4] DeVreese, L., “New Year's Eve Tragedy in Antwerp Fire,” Vol. 89, No. 1092, pp. 4-5,
June 1996.

[5] Brannigan, F. L., “Beware the Christmas Tree and Other Hazards,” Fire Engineering,
Vol. 149, No. 10, pp. 108-110, October 1996.

[6] “CPSC Releases Holiday Safety Tips for Avoiding Fires and Injuries,” Press Release No.
99-029, U.S. Consumer Product Safety Commission, Washington, DC, November 30,
1998.

[7] Chaney, J., “New Regulations Allow Live Holiday Trees in Offices,” Gaithersburg
Gazette, Gaithersburg, MD, December 16, 1998.

[8] Peacock, R.D., J.N. Breese, and C.L. Forney, "A Users Guide for RAPID, Version 2.3,"
NIST Special Publication 798, National Institute of Standards and Technology,
Gaithersburg, MD, January 1991.

[9] Parker, W., "Calculations of the Heat Release Rate by Oxygen Consumption for Various
Applications," NBSIR 81-2427, National Bureau of Standards, Gaithersburg, MD, March
1982.

[10] Parker, W., "Calculations of the Heat Release Rate by Oxygen Consumption for Various
Applications," J. of Fire Sciences, 2, September/October 1984, pp. 380-395, 1982.

[11] Janssens, M.L., "Measuring Rate of Heat Release by Oxygen Consumption," Fire
Technol., 27, pp. 234-249, 1991.

[12] Stroup, D.W., et. al., "Large Fire Research Facility (Building 205) Heat Release Rate
Measurement System," National Institute of Standards and Technology, Gaithersburg,
MD, to be published.

[13] "Design and Manufacture of Medical and Thermal Instrumentation," Medtherm
Corporation, Huntsville, AL, July 1996.

Click To Enlarge
Figure 2. Photograph Showing Two Forms of an Electric Match, One with a Single
Match and One with a Matchbook

Click To Enlarge
Figure 3. Photograph Showing Placement of an Electric Match in the Branches of a Tree

Click To Enlarge

Figure 4. Graph of Heat Release Rate versus Time for Test No. 1

Click To Enlarge
Figure 5. Graph of Heat Release Rate versus Time for Test No. 2

Click To Enlarge
Figure 6. Graph of Heat Release Rate versus Time for Test No. 3

Click To Enlarge
Figure 7. Graph of Heat Release Rate versus Time for Test No. 4

Click To Enlarge
Figure 8. Graph of Heat Release Rate versus Time for Test No. 5

Click To Enlarge
Figure 9. Graph of Heat Release Rate versus Time for Test No. 6

Click To Enlarge
Figure 10. Graph of Heat Release Rate versus Time for Test No. 7

Click To Enlarge
Figure 11. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 1

Click To Enlarge
Figure 12. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 2

Click To Enlarge
Figure 13. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 3

Click To Enlarge

Figure 14. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 4

Click To Enlarge
Figure 15. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 5

Click To Enlarge
Figure 16. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 6

Click To Enlarge
Figure 17. Graph of Total Heat Flux (solid line) and Radiative Heat Flux (dashed line)
versus Time for Test No. 7

Click To Enlarge
Figure 18. Photograph of Tree Prior to Start of Test No. 2 with Arrow Indicating
Ignition Location

Click To Enlarge
Figure 19. Photograph of Tree at Peak Heat Release Rate During Test No. 2

Click To Enlarge
Figure 20. Photograph of Tree Prior to Start of Test No. 3 with Arrow
Indicating Ignition Location

Click To Enlarge
Figure 21. Photograph of Tree at Peak Heat Release Rate During Test No. 3

Click To Enlarge
Figure 22. Photograph of Tree Prior to Start of Test No. 4 with Arrow Indicating
Ignition Location

Click To Enlarge
Figure 23. Photograph of Tree at Peak Heat Release Rate During Test No. 4

Click To Enlarge
Figure 24. Photograph of Tree Prior to Start of Test No. 5 with Arrow
Indicating Ignition Location

Click To Enlarge
Figure 25. Photograph of Tree at Peak Heat Release Rate During Test No. 5

Click To Enlarge
Figure 26. Photograph of Tree Prior to Start of Test No. 6 with Arrow Indicating
Ignition Location

Click To Enlarge
Figure 27. Photograph of Tree at Peak Heat Release Rate During Test No. 6

Click To Enlarge
Figure 28. Photograph of Tree Prior to Start of Test No. 7 with Arrow Indicating
Ignition Location

Click To Enlarge
Figure 29. Photograph of Tree at Peak Heat Release Rate During Test No. 7

Click To Enlarge
Figure 30. Photograph of Tree Prior to Start of Test No. 8 with Arrow Indicating
Ignition Location

Click To Enlarge
Figure 31. Photograph of Tree After Ignition Attempt During Test No. 8