A critical appraisal of charcoal morphometry as a paleofire fuel type proxy

Abstract

Fire is an important disturbance mechanism that can alter ecosystems and catalyze vegetation changes. Paleofire records provide a means of inferring how ecosystems will be impacted by fire regime changes responding to anthropogenic climate change. Although sedimentary charcoal is a reliable fire proxy, recent advances have facilitated more nuanced reconstruction of fire regime changes. Charcoal morphometry, the measurement of sedimentary charcoal particle shape attributes, has emerged as a promising method of differentiating fuel types burned from paleofire records. In particular, the length:width (L:W) ratio of particles has been shown to be a reliable means of distinguishing some fuel types. However, relatively few studies have characterized the L:W values of charcoal using experimental samples, limiting our understanding of this fuel type proxy and its potential applicability in varied environmental settings. We compiled the most comprehensive dataset of experimental charcoal L:W measurements to date and evaluated these data to determine the reliability of L:W measurements as a proxy of the fuel type burned. We founnd that different plant types and plants with different growth habits tend to have significantly different L:W values, in support of previous studies but with some additional nuances. We assessed previously published interpretations of sedimentary charcoal L:W values and show that the bulk of these conclusion are unsupported by our compiled dataset, which could be due to the interpretation of charcoal mixtures of unknown relative amounts of different fuel types. Many of these studies made conflicting interpretations regarding the differentiation of woody and non-woody fuels. In light of our compiled dataset, we propose interpreting L:W values greater than ca. 3.5 to signify grass, non-woody fuels, and values less than ca. 2.5 to signify charcoal derived from woody sources. Similarly, an assessment of fuel type interpretation associated with the L:W values of a published morphological key shows that many of these inferences are also unsupported by the aggregated experimental data, underscoring the need for inclusion of experimental approaches with the creation of morphological keys. Our comparison of plant phylogeny with charcoal L:W values indicated that grass (monocot) fuels may produce more elongate charcoal as a result of evolution-driven differences in plant chemical vascular tissues and morphology. Finally, we summarize the gaps in experimental studies and outline areas of future research needed to reinforce L:W as a reliable fuel type proxy.