Moisture content ranges in live Eucalyptus leaves vary among species and strongly affect ignitability

Abstract Background Although the extent and severity of fires is linked to moisture content in live and dead plant material, the ranges of moisture content occurring in live foliage are poorly defined. The biogeographical distribution of morphological ( e.g. , leaf mass per area) and physiological traits ( e.g. , water potential at turgor loss point) leads us to expect that ranges of moisture content in live foliage are similarly linked to climate, with lower ranges in species from more arid environments. We used pressure–volume and hydraulic vulnerability curves combined with measurements of leaf mass and leaf area to estimate the ranges of moisture content in leaves of some of the dominant Eucalyptus species in south-east Australia, from saturation to hydraulic failure. We also manipulated the moisture content of leaves of the study species by wetting and drying individual leaves which were then exposed to radiant heat to test time to ignition and combustion time across ranges of moisture content. Results The study species occupied different ranges of moisture content, which were limited by leaf saturation and the estimated point of hydraulic failure; saturated moisture contents ranged from 108–152%, and lethal moisture contents ranged from 59–80%. Species with more humid climatic ranges had higher ranges of moisture content. Higher moisture content was found to linearly increase time to ignition in leaves of all species. Time to ignition at the species-specific saturated and lethal moisture content, and the rate of change of time to ignition against moisture content, varied among species according to leaf mass per area, with leaves from species with higher leaf mass per area being slower to ignite. Conclusions We quantified ranges of moisture contents in live leaves and identified traits which could predict how they vary with environmental gradients. Leaf mass per area strongly affected time to ignition but appears to have counter-acting influences on ignitability by simultaneously affecting moisture content and rate of heating. While moisture content increased time to ignition, we found no evidence of a critical threshold of moisture content that determined whether leaves ignited. Models of live fuel moisture content could be constrained by plant traits; some, such as leaf mass per area, are easy to measure, although others, such as hydraulic vulnerability thresholds, are not.