Black carbon (BC, i.e. charcoal) has long been considered an inert and highly resistant component of the soil C pool, explaining its persistence in some soils for thousands of years. Because of its relative recalcitrance, even small concentrations of BC can be significant as a long-term soil C sink. However, recent evidence suggests that BC is retained unequally across ecosystems. Some studies of climatic gradients found that wetter areas with less frequent fires had larger pools of BC because of incomplete combustion of organic matter. In addition, historical land use may influence long-term BC retention. The principal objectives of this study were to identify patterns and mechanisms driving retention of BC across ecosystem gradients, and to explore the influence of historical land use. Soils were collected from a precipitation gradient in the Hawaiian Islands, at sites where invasive grasses have driven a recent increase in wildfire activity. The study site is also an area of known ancient Polynesian agricultural activity. Soils were collected and analyzed for chemical characteristics along the precipitation gradient, and from under archeological walls at paired sites. Soils from under walls were used to explore potential inputs of BC from historical land use.

Results/Conclusions

Chemical analysis (¹³C NMR) showed that concentrations of BC in soils were negatively correlated with precipitation from 800 to 1600 mm MAP (R² = 0.74, p < 0.05, n = 9), with similar trends in fields and under walls. Although BC concentrations declined with precipitation, radiocarbon dating of BC under walls showed a positive correlation between precipitation and BC age (R² = 0.57, p < 0.05, n = 15). Calculated ages of BC under walls indicated inputs during the period of Polynesian agricultural activity (~200 to 700 years ago). The relationship of BC age with precipitation could indicate either a positive effect of rainfall on the duration of BC retention, or historical patterns in the expansion of Polynesian agricultural activity. Black C represented up to 10 % of soil C in fields, and up to 15 % of C under archaeological walls. Because of larger overall C pools in fields, stocks of BC were larger in fields (4.5 mg BC/g soil) than under walls (2 mg BC/g soil). Together, these results indicate that climate is likely a driver of BC stocks in this ecosystem. Additionally, there is an apparent ancient input of BC from Polynesian agricultural practices to these soils.