COS 80-7 - Leaf traits and litter flammability: Understanding multi-species mixtures

Wednesday, August 10, 2011: 3:40 PM
13, Austin Convention Center
Rita Margarida Quinones-Magalhaes and Dylan Schwilk, Biological Sciences, Texas Tech University, Lubbock, TX

Past research has demonstrated the effect of community composition on fire regimes. However, we currently lack a sound mechanistic view of how plant traits influence flammability. A more mechanistic perspective is particularly important as managers attempt to understand and predict fire in plant communities that are changing in response to climate change or altered fire regimes. The effect of a species on the flammability of a community might be non-additive or disproportionate to its abundance. Understanding possible non-additive effects is essential for predicting flammability of novel species combinations based on leaf traits.

This work addresses three questions: 1) How do 8 species common in Sierra Nevada mixed conifer forest differ in their litter flammability?; 2) What leaf traits are associated with various flammability components?; and 3) Do individual species measurements predict multi-species combinations or are there non-additive effects?

Leaf litter material was collected in Sequoia and Kings Canyon National Parks, California, from 8 common species found in mixed-conifer forest: Pinus jeffreyi, P. lambertiana, P. ponderosa, Abies concolor, A. magnifica, Calocedrus decurrens, Quercus kelloggii and Sequoiadendron giganteum. The flammability tests were performed both on monospecific litter beds (5 replicates each) and on litter beds composed of three species at a time (3-5 replicates). We measured flammability parameters (ignitability, sustainability, spread rate, maximal flame height, consumability and heat release) on a 150 cm long burn table instrumented with nine thermocouples.

We tested for non-additive effects in multi-species mixtures by creating a null model using a simple weighted average of single species measures for each flammability component as a null expectation for each mixture and departures from this null to detect non-additive effects.


Results from single-species burn trials show consistent differences between species across most of the different components of flammability. Additionally. most flammability components show a departure from the null expectation, indicating an overall presence of non-additive effects. The exceptions are total heat release and peak temperature, which show a simple additive effect. Flame spread rate is tightly correlated with litter density, which is predicted by leaf size. Other flammability measurements are not closely linked with litter density, however. Understanding the burning characteristics of mixed stands will prove valuable in classifying species in terms of wildfire danger and aid prescribed burning plans, considering the flammability differences of the species.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.