Biomass burning emission analysis based on MODIS aerosol optical depth and AeroCom multi-model simulations: Implications for model constraints and emission inventories
Petrenko M., Kahn R., Chin M., Bauer SE., Bergman T., Bian H., Curci G., Johnson B., Kaiser JW., Kipling Z., Kokkola H., Liu X., Mezuman K., Mielonen T., Myhre G., Pan X., Protonotariou A., Remy S., Skeie RB., Stier P., Takemura T., Tsigaridis K., Wang H., Watson-Parris D., Zhang K.
We assessed the biomass burning (BB) smoke aerosol optical depth (AOD) simulations of 11 global models that participated in the AeroCom phase III BB emission experiment. By comparing multi-model simulations and satellite observations in the vicinity of fires over 13 regions globally, we (1) assess model-simulated BB AOD performance as an indication of smoke source-strength, (2) identify regions where the common emission dataset used by the models might underestimate or overestimate smoke sources, and (3) assess model diversity and identify underlying causes as much as possible. Using satellite-derived AOD snapshots to constrain source strength works best where BB smoke from active sources dominates background non-BB aerosol, such as in boreal forest regions and over South America and southern hemispheric Africa. The comparison is inconclusive where the total AOD is low, as in many agricultural burning areas, and where the background is high, such as parts of India and China. Many inter-model BB AOD differences can be traced to differences in values for the mass ratio of organic aerosol to organic carbon, the BB aerosol mass extinction efficiency, and the aerosol loss rate from each model. The results point to a need for increased numbers of available BB cases for study in some regions and especially to a need for more extensive regional-to-global-scale measurements of aerosol loss rates and of detailed particle microphysical and optical properties; this would both better constrain models and help distinguish BB from other aerosol types in satellite retrievals. More generally, there is the need for additional efforts at constraining aerosol source strength and other model attributes with multi-platform observations.