Dark brown carbon and aerosol mixing impacts on absorbing aerosol optical depth in the southeastern United States
Abstract. Wildfires are becoming more prevalent in the southeastern United States. Wildfires emit light absorbing organic aerosols such as black carbon and dark brown carbon (tar balls) that influence regional air quality and radiative forcing. Recent experimental work has identified dark brown carbon as a potentially significant source of warming in the atmosphere. However, dark brown carbon is not typically included in air quality and climate models. We added dark brown carbon from wildfires as a species in the Weather Research and Forecasting modeling system coupled with chemistry (WRF-Chem) version 4.2 with the Quick-Fire Emissions Database (QFED). The model simulations were run over the southeastern United States from June 21st to June 30th 2013 in conjunction with in-situ measurements of aerosol extinction from the Southeast Nexus (SENEX) aircraft campaign and aerosol optical depth (AOD) measurements from the ground-based remote sensing Aerosol Robotic Network (AERONET). The impacts of imaginary refractive index, aerosol mixing, and Mie method were investigated with an offline radiation model for computationally efficient evaluation. Dark brown carbon was found to increase absorbing aerosol optical depth (AAOD) by up to 5.2 % in the simulation time and region. We derived a linear equation to estimate the correlation between modeled AAOD and columnar dark brown carbon (AAOD = 0.0023[D-BrC] + 0.0007) with core-shell considerations. Core-shell aerosol mixing and explicit Mie computation enhanced increases in AAOD from dark brown carbon at AERONET sites (up to 6.42 %). These results indicate that models omitting dark brown carbon may underestimate aerosol absorption in wildfire-influenced regions, and that including it, along with a realistic mixing state and Mie treatment, is important for estimating radiative forcing as wildfire activity increases.
This manuscript by Ng et al. implemented a dark brown carbon representation in the WRF-Chem model and investigated its implications for aerosol absorption in southeastern US. The topic is important to the community; however, important methodology description and model evaluations are not clearly presented in the present manuscript, making the conclusions hard to evaluate. I would suggest improvements focusing on the following points.
Major concerns:
For 3.3, why do you transfer the AEROENET 440nm AAOD to 550nm? BrC will absorb more at lower wavelength so 440nm would be a better choice for your model evaluation.
For 3.4, I do not think any comparison to extension is relevant to your topic here.
Other comments:
- Abstract: provide the wavelength for AAOD
- Citation and reference: please make the format consistent
- Section 2.3: as you later use AERONET AAOD, you need to introduce the details for measured SSA. Btw, data availability for AERONET Level 2.0 SSA is poor, potentially making the mean AAOD during a period biased high, how did you deal with that?
- Section 2.4 should be clearer. What are the wavelength dependence of refractive indices and size distribution of BC/BrC in the model?
- Section 2.6: when saying include dark brown carbon in the core, do you mean add the dark BrC mass to the BC core, or include dark BrC as a separated core externally mixed with BC?
- Fig 2: the caption says total column, but with a unit of µg m-3?
- Section 3.5: I do not think I demonstrate any relationship, as your added dark BrC contribute too small to 550nm AAOD. Your x-axis is largely correlated with biomass burning emissions, so the linear relationship here in fact tells you the AAOD is related to biomass burning BC.
- Fig 5: the colorbar does not make sense as there is nearly only one color in your plot.
- 3.6 Better to quantify the uncerntaities.