Aerosol swelling: A major source of error when estimating the aerosol first indirect effect



Li, Zhanqing - University of Maryland

Area of research:

Cloud-Aerosol-Precipitation Interactions

Journal Reference:

Liu J and Z Li. 2018. "Significant Underestimation in the Optically Based Estimation of the Aerosol First Indirect Effect Induced by the Aerosol Swelling Effect." Geophysical Research Letters, 45(11), 10.1029/2018GL077679.


There are few measurements of the concentration of cloud condensation nuclei (CCN), so aerosol optical depth (AOD) is widely used as a proxy when estimating the aerosol first indirect effect (FIE), the foundation of aerosol-cloud interactions in numerous forms. Unlike CCN, AOD may be significantly influenced by the aerosol swelling effect. This can lead to uncertainties in the estimation of the FIE and may explain why there is such a large range of values for FIE estimated from observations.


We identify a major source of error in the estimation of the FIE, namely, the aerosol swelling effect, that if accounted for, could narrow the range of FIE observation-based and model-simulated estimates. This would bring FIE estimates more into line with what is theoretically expected.


We used extensive measurements of aerosol and cloud properties made at four Atmospheric Radiation Measurement sites around the world to identify and quantify the influence of aerosol hygroscopicity and the swelling effect on the aerosol FIE. These sites have distinct aerosol properties and experience different meteorological conditions. The magnitude of the FIE for aerosol particles with stronger aerosol hygroscopicity is systematically larger than that for aerosol particles with weaker aerosol hygroscopicity. A one-unit enhancement in the aerosol scattering coefficient by the swelling effect leads to a systematic underestimation of the FIE by about 23%. This can result in a significant underestimation of the FIE-related radiative forcing (by several W m-2 depending on aerosol properties and relative humidity). This likely contributes significantly to the systematic difference between observation-based, especially satellite-based, estimates of the FIE and those simulated by general circulation models. It may also partially explain systematic variations in the FIE with water vapor amount as noted by others.