Investigating FTIR spectroscopy and boron isotopes as novel proxies for wildfire characteristics

Wildfires are a critical component of the natural environment that controls landscape and biological evolution. Our ability to model how fire regimes may change in response to climate change is therefore crucial to understanding and managing the effects of wildfires. In Australia, fire records extend back 60 years at most, which is too short to construct a reliable fire regime model. Ongoing efforts to extend the fire record using proxies such as charcoals are hindered by the difficulty in constraining important fire characteristics such as fire severity and fire intensity, restricting our understanding of the environmental response to wildfires in the past. Thus, there is an urgent need
for new methods of constraining fire characteristics of past wildfires. Boron isotopes are cycled heavily through vegetation and could potentially be used to infer fire severity. However, boron isotope fractionation is sensitive to many factors, and their behavior during wildfires has not been analysed. In addition, previous studies have shown FTIR spectroscopy of soils and charcoals to be sensitive to artificial heating, but the sensitivity has never been tested with wildfires in the natural environment. This study investigates how wildfires modify boron isotopes and FTIR spectra of soils and charcoals, thereby testing their viability as novel fire characteristic proxies. Soil and charcoal samples from Yengo National Park and the Blue Mountains (NSW, Australia), which had last experienced a fire in 2001/2002 and 2019-2020 fire seasons, respectively, were analysed. In addition, lithium isotopes, which are not cycled through vegetation but otherwise behaves similarly to boron isotopes, were analysed for soils from Yengo National Park, the Blue Mountains, and Namadgi National Park (ACT, Australia), in order to elucidate the effects of wildfires on abiotic systems. Boron isotopes are highly fractionated by plants, resulting in leaves being isotopically heavier than barks. This intra-plant boron isotope fractionation is the dominant factor controlling boron isotope composition in soils following wildfires. Higher severity fires combust more leaves, imparting a heavier boron isotope composition to soils relative to low severity fires, as observed in Yengo National Park. However, soil samples from the Blue Mountains suggest a time lag of more than two years may be required to increase the hysteresis of boron isotope signals in the soil, so that boron adsorbed onto clays is not easily desorbed by water. In contrast, bark charcoals created in higher severity fires have a lighter boron isotope composition. As the intra-plant boron isotope fractionation is not involved, the boron isotope compositions of charcoals are likely controlled by combustion temperature and boron isotope fractionation during volatilisation. Therefore, boron isotopes in both soil and charcoal can respond to fire severity, although boron isotopes in charcoals are more reliable, and can differentiate moderate or higher severity fires from low severity fires with a 85 % accuracy. FTIR spectroscopy shows that in both charcoals and soils, the aromatic/aliphatic peak area ratios are higher for higher severity fires, due to the thermal destruction of aliphatic compounds. However, necromass deposition in the soil could disrupt this trend in the soil. The dehydroxylation of clay minerals such as kaolinite and gibbsite can also be detected with FTIR spectroscopy, which helps constrain soil temperature during different severity fires. Both the aromatic/aliphatic peak
area ratio and clay alteration are sensitive to temperature, thus allowing FTIR spectroscopy to infer fire intensity of wildfires. Lithium isotopes in the soil have large variations depending on the isotope composition of parent materials. After experiencing a fire of moderate or higher severity, lithium isotope compositions of soils converge to 0 h. How wildfires may homogenise soil lithium isotopes is unknown, with lithium input from plant and aeolian sources or output from leaching being inadequate explanations. The sensitivity of lithium isotopes to fire severity suggest their use as a robust fire severity proxy, and future research should focus on understanding the mechanism of their response to wildfires. Overall, this study highlights the potential use of boron and lithium isotopes, in addition to FTIR spectroscopy, as novel fire severity or intensity proxies.