Charcoal morphologies and morphometrics of a Eurasian grass-dominated system for robust interpretation of past fuel and fire type

Abstract. Recent developments in morphological and morphometric analyses of charcoal particles have improved our ability to discern characteristics of burnt plant fuel and interpret fire-type changes. However, burning experiments linking known plants to these metrics are limited, particularly in open ecosystems. This study presents novel analyses of laboratory-produced charcoal of 22 plant species from the steppe regions of Eurasia (Romania and Russia), along with selected samples from three Holocene charcoal and pollen records from the same areas. We characterise charcoal production, morphologies and morphometrics in these grass-dominated environments, thereby enabling more robust interpretations of fuel sources and fire types for palaeofire research. Our experiments demonstrate that fire temperature can introduce biases in charcoal produced among species. Grass charcoal production was significantly lower and decreased more strongly with fire temperature compared to forbs. This suggests an underrepresentation of terrestrial graminoids in sedimentary charcoal assemblages. Morphometric analyses revealed that graminoid charcoal particles were more elongated (length-to-width ratio L/W=4) and narrower (width-to-length ratio W/L=0.38) than forbs (L/W=3.1 and W/L=0.42, respectively), in agreement with a global compilation for graminoids (L/W=4.3 for grass 5.4 grass and wetland graminoids) and forbs (L/W=2.9). However, overlapping L/W values present a challenge for establishing cut-off values for fuel type identification in charcoal assemblages with mixed fuel sources. Based on our analyses and compiled datasets from experimental burns, L/W values above 3.0 may indicate predominantly herbaceous morphologies in temperate grassland-dominated ecosystems, though values are likely to be higher for grass than forb-dominated grasslands. Notably, terrestrial grasses exhibit shorter aspect ratios (L/W=4.3) than wetland graminoids (L/W=6.4), highlighting that the aspect ratio needs tailoring to the specific environment of its application, i.e. wetland vs. terrestrial ecosystems. The long forms of graminoid charcoal particles also suggest their potential for atmospheric longer-distance transport compared to more spherical particles, meaning they likely provide insights into regional fire history. An important finding is that charcoal of herbaceous plants closely corresponded to the pollen record, highlighting a solid link between the dominant vegetation and fuel burnt in grassland-dominated environments. However, the relationship between woody charcoal and tree pollen may be more complex, as tree pollen can travel atmospherically longer distances compared to woody charcoal. Our results also highlight the complex interplay between local vegetation and charcoal composition with human fire use that needs to be considered when interpreting charcoal morphological records. A critical takeaway from this study is the importance of not assuming the universality of previous research findings and instead employing experimental approaches to characterise charcoal particles in new ecosystems prior to the application of these techniques. Furthermore, this study also highlights recommendations for further research in new geographical areas and proposes methodological adjustments to enhance the usefulness of charcoal analysis in fire research.