Advanced

Modeling the interactions between aerosols and liquid water clouds with a self-consistent cloud scheme in a general circulation model

Ming, Yi; Ramaswamy, V.; Donner, Leo J.; Phillips, Vaughan LU ; Klein, Stephen A.; Ginoux, Paul A. and Horowitz, Larry W. (2007) In Journal of Atmospheric Sciences 64(4). p.1189-1209
Abstract
To model aerosol-cloud interactions in general circulation models (GCMs), a prognostic cloud scheme of cloud liquid water and amount is expanded to include droplet number concentration (N(d)) in a way that allows them to be calculated using the same large-scale and convective updraft velocity field. In the scheme, the evolution of droplets fully interacts with the model meteorology. An explicit treatment of cloud condensation nuclei (CCN) activation enables the scheme to take into account the contributions to N(d) of multiple aerosol species (i.e., sulfate, organic, and sea-salt aerosols) and to consider kinetic limitations of the activation process. An implementation of the prognostic scheme in the Geophysical Fluid Dynamics Laboratory... (More)
To model aerosol-cloud interactions in general circulation models (GCMs), a prognostic cloud scheme of cloud liquid water and amount is expanded to include droplet number concentration (N(d)) in a way that allows them to be calculated using the same large-scale and convective updraft velocity field. In the scheme, the evolution of droplets fully interacts with the model meteorology. An explicit treatment of cloud condensation nuclei (CCN) activation enables the scheme to take into account the contributions to N(d) of multiple aerosol species (i.e., sulfate, organic, and sea-salt aerosols) and to consider kinetic limitations of the activation process. An implementation of the prognostic scheme in the Geophysical Fluid Dynamics Laboratory (GFDL) AM2 GCM yields a vertical distribution of N(d) with a characteristic maximum in the lower troposphere; this feature differs from the profile that would be obtained if N(d) is diagnosed from the sulfate mass concentration based on an often-used empirical relationship. Prognosticated N(d) exhibits large variations with respect to the sulfate mass concentration. The mean values are generally consistent with the empirical relationship over ocean, but show negative biases over the Northern Hemisphere midlatitude land, perhaps owing to the neglect of subgrid variations of large-scale ascents and inadequate convective sources. The prognostic scheme leads to a substantial improvement in the agreement of model-predicted present-day liquid water path (LWP) and cloud forcing with satellite measurements compared to using the empirical relationship. The simulations with preindustrial and present-day aerosols show that the combined first and second indirect effects of anthropogenic sulfate and organic aerosols give rise to a steady-state global annual mean flux change of -1.8 W m(-2), consisting of -2.0 W m(-2) in shortwave and 0.2 W m(-2) in longwave. The ratios of the flux changes in the Northern Hemisphere (NH) to that in Southern Hemisphere (SH) and of the flux changes over ocean to that over land are 2.9 and 0.73, respectively. These estimates are consistent with the averages of values from previous studies stated in a recent review. The model response to higher N(d) alters the cloud field; LWP and total cloud amount increase by 19% and 0.6%, respectively. Largely owing to high sulfate concentrations from fossil fuel burning, the NH midlatitude land and oceans experience strong radiative cooling. So does the tropical land, which is dominated by biomass burning-derived organic aerosol. The computed annual, zonal-mean flux changes are determined to be statistically significant, exceeding the model's natural variations in the NH low and midlatitudes and in the SH low latitudes. This study reaffirms the major role of sulfate in providing CCN for cloud formation. (Less)
Please use this url to cite or link to this publication:
author
publishing date
type
Contribution to journal
publication status
published
subject
in
Journal of Atmospheric Sciences
volume
64
issue
4
pages
1189 - 1209
publisher
Amer Meteorological Soc
external identifiers
  • wos:000245742600009
  • scopus:34247554922
ISSN
1520-0469
DOI
10.1175/JAS3874.1
language
English
LU publication?
no
id
c92c05fa-8efc-44fb-8c8d-5ee4886db496 (old id 4587521)
date added to LUP
2014-08-15 10:19:38
date last changed
2017-11-05 03:39:15
@article{c92c05fa-8efc-44fb-8c8d-5ee4886db496,
  abstract     = {To model aerosol-cloud interactions in general circulation models (GCMs), a prognostic cloud scheme of cloud liquid water and amount is expanded to include droplet number concentration (N(d)) in a way that allows them to be calculated using the same large-scale and convective updraft velocity field. In the scheme, the evolution of droplets fully interacts with the model meteorology. An explicit treatment of cloud condensation nuclei (CCN) activation enables the scheme to take into account the contributions to N(d) of multiple aerosol species (i.e., sulfate, organic, and sea-salt aerosols) and to consider kinetic limitations of the activation process. An implementation of the prognostic scheme in the Geophysical Fluid Dynamics Laboratory (GFDL) AM2 GCM yields a vertical distribution of N(d) with a characteristic maximum in the lower troposphere; this feature differs from the profile that would be obtained if N(d) is diagnosed from the sulfate mass concentration based on an often-used empirical relationship. Prognosticated N(d) exhibits large variations with respect to the sulfate mass concentration. The mean values are generally consistent with the empirical relationship over ocean, but show negative biases over the Northern Hemisphere midlatitude land, perhaps owing to the neglect of subgrid variations of large-scale ascents and inadequate convective sources. The prognostic scheme leads to a substantial improvement in the agreement of model-predicted present-day liquid water path (LWP) and cloud forcing with satellite measurements compared to using the empirical relationship. The simulations with preindustrial and present-day aerosols show that the combined first and second indirect effects of anthropogenic sulfate and organic aerosols give rise to a steady-state global annual mean flux change of -1.8 W m(-2), consisting of -2.0 W m(-2) in shortwave and 0.2 W m(-2) in longwave. The ratios of the flux changes in the Northern Hemisphere (NH) to that in Southern Hemisphere (SH) and of the flux changes over ocean to that over land are 2.9 and 0.73, respectively. These estimates are consistent with the averages of values from previous studies stated in a recent review. The model response to higher N(d) alters the cloud field; LWP and total cloud amount increase by 19% and 0.6%, respectively. Largely owing to high sulfate concentrations from fossil fuel burning, the NH midlatitude land and oceans experience strong radiative cooling. So does the tropical land, which is dominated by biomass burning-derived organic aerosol. The computed annual, zonal-mean flux changes are determined to be statistically significant, exceeding the model's natural variations in the NH low and midlatitudes and in the SH low latitudes. This study reaffirms the major role of sulfate in providing CCN for cloud formation.},
  author       = {Ming, Yi and Ramaswamy, V. and Donner, Leo J. and Phillips, Vaughan and Klein, Stephen A. and Ginoux, Paul A. and Horowitz, Larry W.},
  issn         = {1520-0469},
  language     = {eng},
  number       = {4},
  pages        = {1189--1209},
  publisher    = {Amer Meteorological Soc},
  series       = {Journal of Atmospheric Sciences},
  title        = {Modeling the interactions between aerosols and liquid water clouds with a self-consistent cloud scheme in a general circulation model},
  url          = {http://dx.doi.org/10.1175/JAS3874.1},
  volume       = {64},
  year         = {2007},
}