The global aerosol‐cloud first indirect effect estimated using MODIS, MERRA, and AeroCom

Abstract

Aerosol‐cloud interactions (ACI) represent a significant source of forcing uncertainty in global climate models (GCMs). Estimates of radiative forcing due to ACI in Fifth Assessment Report range from −0.5 to −2.5 W m−2. A portion of this uncertainty is related to the first indirect, or Twomey, effect whereby aerosols act as nuclei for cloud droplets to condense upon. At constant liquid water content this increases the number of cloud droplets (Nd) and thus increases the cloud albedo. In this study we use remote‐sensing estimates of Nd within stratocumulus regions in combination with state‐of‐the‐art aerosol reanalysis from Modern‐Era Retrospective Analysis for Research and Applications version 2 (MERRA2) to diagnose how aerosols affect Nd. As in previous studies, Nd is related to sulfate mass through a power law relationship. The slope of the log‐log relationship between Nd and SO4 in maritime stratocumulus is found to be 0.31, which is similar to the range of 0.2–0.8 from previous in situ studies and remote‐sensing studies in the pristine Southern Ocean. Using preindustrial emissions models, the change in Nd between preindustrial and present day is estimated. Nd is inferred to have more than tripled in some regions. Cloud properties from Moderate Resolution Imaging Spectroradiometer (MODIS) are used to estimate the radiative forcing due to this change in Nd. The Twomey effect operating in isolation is estimated to create a radiative forcing of −0.97 ± 0.23 W m−2 relative to the preindustrial era.

Predicting decadal trends in cloud droplet number concentration using reanalysis and satellite data

“…The analysis performed in this study parallels the analysis in McCoy et al (2017a). Here a much more refined data set is used to analyze the period 2003(as opposed to 2001in McCoy et al (2017a), expanded to a daily time 15 scale over the entire globe (21,379,174 daily 1°x1° observations).…”

“…The aerosol species considered in the present analysis are dust (DU), 20 sea salt (SS), black carbon (BC), organic carbon (OC) and sulfate (SO 4 ). As in McCoy et al (2017a), DU and SS masses as predicted by MERRA2 are restricted to submicron sizes because these will be more numerous as CCN (Ghan et al, 1998).…”

“…Here a much more refined data set is used to analyze the period 2003(as opposed to 2001in McCoy et al (2017a), expanded to a daily time 15 scale over the entire globe (21,379,174 daily 1°x1° observations). Aerosol reanalysis from MERRA2 is used to gain insight into speciation and vertical distribution that is not provided by remote-sensing analyses that use column-integrated CCN proxies such as aerosol index (AI) or aerosol optical depth (AOD).…”

“…In McCoy et al (2017a) 1°x1° daily-mean MODIS effective radius (r e ) and optical depth (τ) values were used to calculate CDNC, which was then averaged to monthly resolution. The use of this CDNC dataset may be problematic in some regions for a number of reasons (also see McCoy et al (2017a)) : 1) it is subject to high solar zenith angle biases in the individual swaths, which were averaged together to create each daily data point; 2) biases may be present due to the use of area averaged r e and t rather than using pixel level values for the CDNC calculation; 3) the dataset was not filtered to include 30 low altitude clouds only, which may have led to a lack of connectivity between surface aerosol sources and cloud CDNC; 4) the CDNC was calculated using the 2.1µm MODIS channel r e , which is likely to be affected more strongly by cloud Atmos.…”

“…Studies have utilized the natural laboratory provided by transient degassing volcanoes to study cloud responses to changes in aerosol (Mace and Abernathy, 2016;Gassó, 2008;Yuan et al, 2011;McCoy et al, 2017b;Malavelle et al, 2017). In this vein, McCoy et al (2017a) used aerosol reanalysis to provide additional information regarding aerosol speciation and vertical structure. They found that monthly-mean CDNC and sulfate 5 mass concentration near the surface were linked by a power law relationship that remained robust across different regions with very different aerosol properties and cloud regimes, but their analysis was hampered by remote-sensing bias leading to different regions having a different constant term in the log-log fit between CDNC and sulfate (SO 4 ).…”