Reassessment of pre-industrial fire emissions strongly affects anthropogenic aerosol forcing

Uncertainty in pre-industrial natural aerosol emissions is a major component of the overall uncertainty in the radiative forcing of climate. Improved characterisation of natural emissions and their radiative effects can therefore increase the accuracy of global climate and Earth system model projections.

Several lines of evidence have suggested that fire activity was much greater in the pre-industrial era than has previously been assumed. To test the impact of this new understanding we compared fire emissions from two fire models designed to represent historically higher emissions due to land use change to the recent Sixth Coupled Model Intercomparison Project (CMIP6) fire emission dataset. We find emissions in the fire models are between two-and-a-half to five times higher at 1750 compared to CMIP6, resulting in a predicted 35 to 91% reduction in the calculated global mean cloud albedo forcing over the Industrial Era (1750-2000 CE).

Although pre-industrial observational data is sparse, aerosol modelling which incorporates fire model emissions are in closer agreement to paleofire proxy records of ice cores, charcoal data and tree-rings than the model simulation using the CMIP6 dataset. However, CMIP6 is a significant improvement upon the previous 1750 dataset adopted by AeroCom.

It is highly likely that models participating in CMIP6 would simulate smaller anthropogenic aerosol radiative forcings over the industrial period if they used higher fire emissions. Therefore, in order to improve understanding of the magnitude of the historical radiative forcing due to anthropogenic aerosol emissions, as well as other Earth system components such as ozone concentrations and soluble iron deposition, we need to develop a much deeper understanding of how fires evolved over the industrial period, especially since global fire models have mostly been evaluated against present-day observations.