Impact of forest fire on the mercury stable isotope composition in litter and soil in the Amazon

Abstract

Mercury (Hg) emissions from forest fires, especially tropical forests such as the Amazonian forest, were shown to contribute significantly to the atmospheric mercury budget, but new methods are still necessary to improve the traceability and to reduce the great uncertainties related to this emission source. Recent studies have shown that the combustion process can result in Hg stable isotope fractionation that allows tracking coal combustion Hg emissions, as influenced by different factors such as combustion temperature. The main goal of the present study was, therefore, to investigate for the first time the potential of Hg stable isotopes to trace forest fire Hg emissions and pathways. More specifically, small-scale and a large scale prescribed forest fire experiments were conducted in the Brazilian Amazonian forest to study the impact of fire severity on Hg isotopic composition of litter, soil, and ash samples and associated Hg isotope fractionation pathways. In the small-scale experiment, no difference was found in the mercury isotopic composition of the samples collected before and after burning. In contrast, the larger-scale experiment resulted in significant mass dependent fractionation (MDF δ202Hg) in soils and ash suggesting that higher combustion temperature influence Hg isotopic fractionation with the emission of lighter Hg isotopes to the atmosphere and enrichment with heavier Hg in ashes. As for coal combustion, mass independent fractionation was not observed. To our knowledge, these results are the first to highlight the potential of forest fires to cause Hg isotopic fractionation, depending on the fire severity. The results also allowed to establish an isotopic fingerprint for tropical forest fire Hg emissions that corresponds to a mixture of litter and soil Hg isotopic composition (resulting atmospheric δ202Hg, Δ200Hg and Δ199Hg were −1.79 ± 0.24‰, −0.05 ± 0.04‰ and −0.45 ± 0.12‰, respectively).