Why estimates of the peat burned in fires in Sumatra and Kalimantan are unreliable and why it matters

Abstract

Most recent estimates of carbon emissions from Indonesia's peatland fires are based on extrapolation from a narrow base of empirical evidence, raising concerns about the reliability of fire emissions estimates. Measurements of peat fires during ENSO periods are not representative of fires in other years, yet they underlie many estimates of peat fire emissions in ‘normal’ years. Errors may enter into estimates of area burned, quantity of peat combusted and fire emission factors. Problems arise in extrapolating from a few empirical measurements without accounting for conditions that influence combustibility of peat and the heavy fuels through which surface fires transition to become peat fires. These conditions are influenced by drainage and fire history. Our analysis is based on a critical look at a sample of peat fire and emissions studies, including the two most widely cited ones, as well as at the uses made of those studies. Undue extrapolation from unrepresentative empirical studies contributes significantly to the uncertainty of emissions estimates in studies that rely on models rather than empirical observation. We do not offer our own estimates; rather, we argue that the base of evidence must be broadened. We point to examples of empirically based research that more reliably show the contribution of peat fires to overall peatland carbon emissions.