Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Apr 16;45(7):3297-3306.
doi: 10.1002/2018GL077261.

Type-Dependent Responses of Ice Cloud Properties to Aerosols From Satellite Retrievals

Affiliations
Free PMC article

Type-Dependent Responses of Ice Cloud Properties to Aerosols From Satellite Retrievals

Bin Zhao et al. Geophys Res Lett. .
Free PMC article

Abstract

Aerosol-cloud interactions represent one of the largest uncertainties in external forcings on our climate system. Compared with liquid clouds, the observational evidence for the aerosol impact on ice clouds is much more limited and shows conflicting results, partly because the distinct features of different ice cloud and aerosol types were seldom considered. Using 9-year satellite retrievals, we find that, for convection-generated (anvil) ice clouds, cloud optical thickness, cloud thickness, and cloud fraction increase with small-to-moderate aerosol loadings (<0.3 aerosol optical depth) and decrease with further aerosol increase. For in situ formed ice clouds, however, these cloud properties increase monotonically and more sharply with aerosol loadings. An increase in loading of smoke aerosols generally reduces cloud optical thickness of convection-generated ice clouds, while the reverse is true for dust and anthropogenic pollution aerosols. These relationships between different cloud/aerosol types provide valuable constraints on the modeling assessment of aerosol-ice cloud radiative forcing.

Figures

Figure 1.
Figure 1.
Changes in the properties of (a, b) all ice cloud types, (c) anvil ice clouds, (d) in situ ice clouds, and (e) deep convective clouds that anvil ice clouds are connected to, with column/layer aerosol optical depth (AOD). Note that anvil ice clouds are the anvil detrained from deep convective clouds rather than the deep convective core. The deep convective clouds shown in (e) are not the target of this study and are illustrated only to facilitate the interpretation of the aerosol impact on anvil ice clouds. The AOD bins are selected so that each bin contains a similar number of samples. The error bars represent the standard errors (σ/N) where N is the sample number and σ is the standard deviation. The total numbers of samples used in the figures are 2.8 × 105, 5.7 × 104, and 9.8 × 104 for cloud optical thickness, ice cloud fraction (ICF)/ice crystal effective radius (Rei), and ice water path/ice water content of all ice cloud types, 6.2 × 104 and 2.7 × 104 for cloud thickness/cloud optical thickness and ICF/Rei of anvil ice clouds, 2.3 × 105 and 1.1 × 104 for cloud thickness/cloud optical thickness and ICF/Rei of in situice clouds, and 1.7 × 104 for deep convective clouds.
Figure 2.
Figure 2.
Influence of relative humidity on the responses of ice cloud properties to aerosol loadings. (a–c) Changes in (a) cloud thickness, (b) cloud optical thickness (COT), and (c) ice cloud fraction (ICF) with column aerosol optical depth (AOD) under three subsets of relative humidity averaged between 100 and 440 hPa (RH100–440hPa). (d–f) Changes in (d) cloud thickness, (e) COT, and (f) ICF with RH100–440hPa for different AOD ranges. (g–i) Similar to (a)–(c) but for properties of convection-generated (anvil) ice clouds. (j–l) Similar to (a)–(c) but for the changes of the properties of in situ ice clouds with layer AOD. We divide AOD and meteorological variables and into two and three ranges, respectively, each containing a similar sample number. The error bar definition is the same as in Figure 1.
Figure 3.
Figure 3.
The cloud optical thickness (COT) in high and low aerosol optical depth (AOD) subsets for different aerosol types identified by Cloud-Aerosol Lidar with Orthogonal Polarization: (a) all ice cloud types, (b) convection-generated (anvil), and (c) in situ formed ice clouds. Two subsets of column AOD/layer AOD used here contain similar sample number. The differences between the COT under high and low AODs are statistically significant at the 0.01 level based on the Student’s t test except for two cases: For all ice cloud types with smoke aerosols, the difference is significant at the 0.05 level, while for convection-generated ice clouds with anthropogenic pollution, the difference is not significant. The error bar definition is the same as in Figure 1.

Similar articles

See all similar articles

LinkOut - more resources

Feedback