Bushfire occurrences will likely be exacerbated by climate change, thus requiring a model to forecast and manage their impacts. However, a robust bushfire model requires new proxies that can infer fire severity responses to past climate variability. A key test to the viability of new fire proxies is whether they record fire severity in the affected soil. We address this by testing Attenuated Total Reflectance- Fourier Transform Infrared (ATR-FTIR) spectroscopy and boron (B) isotopes in soil clay fractions from Yengo National Park, southeastern Australia, as proxies for bushfire severity. The isotopic results were also compared to that of clays that reacted with experimentally combusted bark. ATR-FTIR spectroscopy constrains the soil temperature to between 225 and 500 ��C during high severity fires, based on the lack of dehydroxylation peak characteristics and the increased aromatic to aliphatic organic peak ratios in clays, compared to that of low severity sites. The isotope composition of the non-exchangeable B fraction in clays is lighter after reacting with leaching solutions of bark combusted at higher temperatures. Combustion temperature does not affect the B isotope fractionation during B adsorption onto clays. Changes to the B isotope composition of clays could instead be justified by the varying B concentration and B isotope compositions of the leaching solutions. In Yengo soil clay fractions, sites that experienced a high severity fire show higher ��11B values by about 1.5 ���, compared to low severity sites- at odds with observations from our experiment using combusted bark. The combustion of leaves from tree crowns in high severity fires could account for the increase in ��11B of clays post-fire. In summary, FTIR spectroscopy of clays could be useful for constraining soil temperature during bushfires, while the B isotope composition of clays appears as a promising proxy for fire severity.