Analyzing PAH Concentrations in Experimental Burn Plot Soils to Improve Reconstructions of Savanna Fires

Today, savanna fires are responsible for 80% of burned areas globally, but it is difficult to predict how these fires may change in the future. To better understand how fires will respond to climate change, scientists reconstruct past changes in savanna fire occurrence using Polycyclic Aromatic Hydrocarbons (PAHs). To do so, we must understand what conditions control the concentrations and distributions of PAHs in savannas. We used soil samples from Kruger National Park, South Africa, where managers have conducted controlled experimental fires for the last 69 years to test how PAHs are preserved in soils with a known fire history. We analyzed the PAH concentrations in soils sampled before and after fires (pre- and post-burn) and in fire suppression plots. PAH concentrations were greater and more variable in burned plots' soil than in unburned plots but did not differ in burn plots pre- vs. post-fire. Unburned plots also had relatively larger and more alkylated PAHs, consistent with degradation of PAHs in soils that have not burned recently. PAH concentrations are higher in plots with greater rainfall, nutrient-rich basalt bedrock, and intermediate fire frequency. Together these variables determine grass biomass, which suggests fuel loads are a major control on PAH production. Our work suggests soil PAH concentrations at local scales are insensitive to individual fires but instead reliably preserve decadal fire history, supporting the use of PAHs to reconstruct fire regimes. However, fuel loads can complicate interpretations of PAH concentrations in fire reconstructions.