Atmospheric Chemistry and Physics (May 2024)

Modeling atmospheric brown carbon in the GISS ModelE Earth system model

  • M. A. DeLessio,
  • M. A. DeLessio,
  • K. Tsigaridis,
  • K. Tsigaridis,
  • S. E. Bauer,
  • S. E. Bauer,
  • J. Chowdhary,
  • J. Chowdhary,
  • G. L. Schuster

DOI
https://doi.org/10.5194/acp-24-6275-2024
Journal volume & issue
Vol. 24
pp. 6275 – 6304

Abstract

Read online

Brown carbon (BrC) is an absorbing organic aerosol (OA), primarily emitted through biomass burning (BB), which exhibits light absorption unique to both black carbon (BC) and other organic aerosols. Despite many field and laboratory studies seeking to constrain BrC properties, the radiative forcing (RF) of BrC is still highly uncertain. To better understand its climate impact, we introduced BrC to the One-Moment Aerosol (OMA) module of the GISS ModelE Earth system model (ESM). We assessed ModelE sensitivity to primary BrC processed through a novel chemical aging scheme and to secondary BrC formed from biogenic volatile organic compounds (BVOCs). Initial results show that BrC typically contributes a top-of-the-atmosphere (TOA) radiative effect of 0.04 W m−2. Sensitivity tests indicate that explicitly simulating BrC (separating it from other OAs), including secondary BrC, and simulating chemical bleaching of BrC contribute distinguishable radiative effects and should be accounted for in BrC schemes. This addition of prognostic BrC to ModelE allows greater physical and chemical complexity in OA representation with no apparent trade-off in model performance, as the evaluation of ModelE aerosol optical depth against Aerosol Robotic Network (AERONET) and Moderate Resolution Imaging Spectroradiometer (MODIS) retrieval data, with and without the BrC scheme, reveals similar skill in both cases. Thus, BrC should be explicitly simulated to allow more physically based chemical composition, which is crucial for more detailed OA studies like comparisons to in situ measurement campaigns. We include a summary of best practices for BrC representation within ModelE at the end of this paper.