1993 Proceedings of the Midwest Oak Savanna Conferences

AN EXAMINATION OF THE HEAT OF COMBUSTION OF DECIDUOUS TREE LEAF LITTER AND HOW IT RELATES TO FIRE ECOLOGY

Several discussions have occurred in the literature dealing with traits of fire-dependent plant communities which confer upon them increased fire frequency and intensity. A study was conducted to determine if the leaves of fire-dependent tree species produce greater amounts of heat when burned than those of non-fire-dependent species. Preliminary results suggest that, within the limits of the heat detection methods used, all species produced similar amounts of heat when burned. A planned refinement of the methods may reveal a finer scale temperature difference between fire-dependent and non-fire-dependent species.

INTRODUCTION 

Mutch (1970) has hypothesized that "if species have developed reproductive mechanisms (underground rhizomes, root sprouting, serotinous cones) and anatomical mechanisms (thick bark, epicormic sprouting) to survive periodic fires, then fire dependent plants might also possess characteristics obtained through natural selection that actually enhance the flammability of these communities," thereby, allowing these fire tolerent species to suppress or eliminate fire intolerant competitors. There has been debate over this hypothesis (Snyder, 1984, Buckley, 1984) as to whether increased flammability is selected for or if it is a secondary effect of other characteristics related to environmental conditions such as moisture loss and herbivory. Flammability of fuels can refer to the ease with which they ignite, the speed at which flames spreads, and the heat produced upon burning. The flammability of a particular genus or species is a relative value depending upon the community in which they are found. Oak and pine are two important forest types that are dependent on natural and anthropogenic fire in North America (Pyne, 1982). In southeastern communities where pine and oak are both present, pines appear to be more flammable than oaks under certain conditions (Platt, et al, 1991). However, in the Midwest, where oak communities often dominate without pines, oaks are thought to possess greater flammability than other deciduous associates (Lorimer, 1987).

Most fire dependent communities are made up of one or a few species responsible for the flammability of the community (Snyder, 1984). In healthy oak communities, the flammability of the site is a combination of the oak leaf litter in addition to several graminoid fuel species. However, in degraded oak woodlands and savannas where the herbaceous layer has been suppressed due to shading, the oak leaf litter and that of non-oak tree species provides most of the fuel.

Oak trees have evolved characteristics which provide them with a greater ability to survive fire. Protection from temperature rise at the cambium layer is dependent only on bark thickness (Vines, 1968). Oaks generally have thicker bark than most other deciduous tree species with the thickest bark occuring in bur oaks which were exposed most frequently to prairie fires (Lorimer, 1987, Curtis, 1969). In addition, most oaks also have the ability to resprout from dormant buds at the base when the top has been killed by fire (Lorimer, 1987). If the hypothesis of Mutch (1970) is valid, and oaks, being a fire adapted genus have selected these adaptations in response to fire, it would then seem likely that increased flammability of oak leaf litter over other deciduous tree species might also be likely. Therefore, it was the purpose of this investigation to examine the heat production of fire tolerant and intolerant deciduous leaf litter to see if fire-tolerant species possess increased flammability in terms of greater heat production as compared to non-tolerant species.

METHODS 

Heat Production - The Fuel
Naturally fallen leaves were collected from seven species of hardwood trees, four fire tolerant species (Quercus bicolor, Q. alba, Q. macrocarpa, and Carya ovata) and three fire intolerant species (Acer saccharum, Fraxinus americana, and Tilia americana)(Curtis, 1969). In the first test with all seven species, the leaves were contained inside plastic trash bags, open at the top, and allowed to air dry inside an unheated steel building for 12 months. In the second test with five of the species (Q. alba, Q. macrocarpa, Q. bicolor, A. saccharum, and F. americana), the leaves were dried inside mesh bags at 53 degrees centigrade for 24 hours in an herbarium plant drier. For both tests, 527 grams of leaves of each species were weighed on an Ohas triple beam balance. The leaves were then spread evenly inside a 1 meter diameter poultry netting enclosure situated on a washed stone substrate in an open area.

Temperature Measurement
Tempil tablets, chemical temperature indicators (available from Tempil, Big Three Industries, Inc., South Plainfield, NJ 07080) with temperature values ranging at intervals from 52 degrees C to 804 degrees C were used to measure fire temperature. In the first test 12 tablets were used in each trial with values ranging from 52 to 804 degrees C. In the second test 5 tablets were used in each trial with values ranging from 302 to 371 degrees C. The tablets were spaced evenly inside a strip constructed from single layers of heavy duty aluminum foil. Platt, et al, (1991) found that aluminum foil raised the ambient temperature at which tablets melted (Y=1.21X42.8). A single aluminum strip containing the tablets was placed on top of the leaf layer and the leaves were then ignited on the upwind side by means of a burning Panama drip torch. After the entire fuel layer was consumed, the strip was removed and the tablets examined. The last tablet either partially or completely melted was considered the temperature of the burn. One to three trials were run on each species in the first test while two to three trials were run in the second test. The first test was conducted under sunny conditions with a high temperature of 69 degrees F, relative humidity of 53 %, and winds of 12 mph with gusts to 21 mph. The second test was conducted on a partly cloudy day with a high temperature of 41 degrees F, relative humidity of 67%, and winds of 15 mph gusting to 20 mph.

RESULTS 

Leaf Burn
The temperature values reported here are the actual tablet ratings and do not take into account the damping effect of the aluminum discussed above. In the first test with all seven species (Fig. 1) it was found that all of the species, with the exception of C. ovata and T. americana, burned at a temperature between 302 and 371 degrees centigrade. In the trials with C. ovata and T. americana the 302 C tablet was partially melted indicating that 302 C was the temperature reached. In the remainder of the species trials the 302 C tablet was completely melted while the 371 C tablet was intact indicating that a temperature midway between these two values was reached. In the second test with five of the above species (Fig. 2) it was found that all of the species burned at 371 C with the exception of F. americana which burned between 343 and 371 C. Most of the species burned consistently at the same temperature. However, there was some variation in two of the species. For F. americana, one of the three trials burned at 343 C while the two other trials were at a value between 343 and 371 C. In Q. macrocarpa, one of the three trials burned at 343 C while the other two trials burned at 371 C. In one of the three Q. alba trials the 371 C tablet was completely melted suggesting that a higher temperature may have been reached.

DISCUSSION 

The results obtained suggest that there has been no selection for increased heat production during the combustion of leaves of the fire dependent species examined. Under the artificial conditions of this experiment, the leaves of all the tree species studied have the potential of burning at temperatures lethal to cambium tissue (60 C 140 F)( Hare, 1965). Cole, et al, (1992) found LD50 temperatures for small trees (2.5 5.0 cm dbh) to be 183 C and for medium trees (5.0 10.0 cm dbh) to be 366 C.

Although all litter samples were burned in the same sized wire enclosures, it is possible that the surface area provided was not sufficient to allow for the chemical reactions of combustion to stabilize before contacting the temperature indicators. Mutch (1970) suggests that fuel chemistry may have a chainlike additive effect when burned. If sufficient time wasn't allowed for these chemical processes to evolve during the burning, maximum energy release may not have been achieved. This could have been tested for by having larger enclosures to allow for longer burn times before contact with the temperature indicators.

While a greater energy release for fire dependent species was not observed, it is likely that other characteristics of fire dependent species leaf litter add to the flammability of these species over nondependent species. The tan in content of oak leaves makes them less palatable to herbivores (Ricklefs, 1973). As a result these leaves may last longer in the environment providing a fuel source over a longer period of time than is the case with some of the non-dependant species such as maple. The open structure of oak communities resulting from fire permits greater insulation and air movement. These drier conditions would also permit oak leaves to persist for longer periods of time. The ability of oak leaves to curl in spring after snow melt is cited as a feature important to flammability (Lorimer, 1987). This ability to curl upon drying in oak leaves is opposed to the limp, matted condition found in maple leaves (Packard, 1987).

CONCLUSION 

Although this experiment suggests that the leaf litter of fire dependent species examined in this study have not selected for greater energy release upon burning, it is believed that the length of time which the leaves of fire dependent species remain as a potential fuel along with other characteristics related to climate and herbivory impart a greater flammability on these species.

Acknowledgments I wish to thank Heather Hagg for assistance while working on this study as part of a science fair project. I also would like to thank David Sollenberger for assistance in conducting the burns and for recording aspects of the study on film.

LITERATURE CITED 

Buckley, R. 1984. The role of fire: response to Snyder. Oikos 43(3):405406.

Cole, K.L., K.F. Klick, and N.B. Pavlovic. 1992. Fire temperature monitoring during experimental burns at Indiana Dunes National Lakeshore. Natural Areas Journal 12(4): 177183.

Curtis, J. 1959. The vegetation of Wisconsin. Univ. of Wisconsin Press, Madison. 657 p.

Hare, R.C. 1965. Contribution of bark to fire resistance. Journal of Forestry 63:248251.

Lorimer, C.G. 1987. The role of fire in the perpetuation of oak forests. Eighth Northern Illinois Prairie Workshop, May, 5975.

Mutch, R. 1970. Wildland fires and ecosystems a hypothesis. Ecology 51: 10461051.

Packard, S. 1987. Rediscovering the tallgrass savanna of Illinois. Eighth Northern Illinois Prairie Workshop, May 2.

Platt, W.J., J.S. Glitzenstein, and D.R. Streng. 1991. Evaluating pyrogenicity and its effects on vegetation in longleaf pine savannas. Proc. Tall Timbers Fire Ecology Conference 17:143161.

Pyne, S.J. 1982. Fire in America: a cultural history of wildland and rural fire. Princeton University Press, 654 p.

Ricklefs, R.E. 1973. Ecology. University of Pennsylvania, 861 p.

Snyder, J.R. 1984. The role of fire: much a do about nothing. Oikos 43(3): 404405.

Vines, R.G. 1968. Heat transfer through bark, and the resistance of trees to fire. Aust. Journal of Botany. 16:499514.

FIGURES