Burn or rot: leaf traits explain why flammability and decomposability are decoupled across species

1. In fireprone ecosystems, two important alternative fates for leaves are burning in a wildfire (when alive or as litter) or they get consumed (as litter) by decomposers. The influence of leaf traits on litter decomposition rate is reasonably well understood. In contrast, less is known about the influence of leaf traits on leaf and litter flammability. The aim of this study was twofold: (i) to determine which morphological and chemical leaf traits drive flammability and (ii) to determine whether different (combinations of) morphological and chemical leaf traits drive interspecific variation in decomposition and litter flammability and, in turn, help us understand the relationship between decomposability and flammability. 2. To explore the relationships between leaf traits and flammability of individual leaves, we used 32 evergreen perennial plant species from eastern Australia in standardized experimental burns on three types of leaf material (i.e. fresh, dried and senesced). Next, we compared these trait-flammability relationships to trait-decomposability relationships as obtained from a previous decomposition experiment (focusing on senesced leaves only). 3. Within the three parameters of leaf flammability that we measured, interspecific variation in time to ignition was mainly explained by specific leaf area and moisture content. Flame duration and smoulder duration were mostly explained by leaf dry mass and to a lesser degree by leaf chemistry, namely, nitrogen, phosphorus and tannin concentrations. 4. The variation in the decomposition constant across species was unrelated to our measures of flammability. Moreover, different combinations of morphological and chemical leaf properties underpinned the interspecific variation in decomposability and flammability. In contrast to litter flammability, decomposability was driven by lignin and phosphorus concentrations. 5. The decoupling of flammability and decomposability leads to three possible scenarios for species' influence on litter fates: (i) fast-decomposing species for which flammability is irrelevant because there will not be enough litter to support a fire; (ii) species with slow-decomposing leaves and a high flammability; and (iii) species with slow-decomposing leaves and a low flammability. We see potential for making use of the decoupled trait-decomposition-flammability relationships when modelling carbon and nutrient fluxes. Including information on leaf traits in models can improve the prediction of fire behaviour. We note that herbivory is another key fate for leaves, but this study was focused on fire and decomposition.