Aerosol–cloud radiative effects are determined and quantified in simulations
of deep open-cell stratocumuli observed during the VAMOS Ocean-Cloud-Atmosphere-Land
Study Regional Experiment (VOCALS-REx) campaign off
the west coast of Chile. The cloud deck forms in a boundary
layer 1.5 km deep, with cell sizes reaching 50 km in diameter. Global databases of ship
tracks suggest that these linear structures are seldom found in boundary
layers this deep. Here, we quantify the changes in cloud radiative properties
to a continuous aerosol point source moving along a fixed emission line
releasing 10
Furthermore, we demonstrate that these changes in cloud radiative properties are masked by the naturally occurring variability within the organised cloud field. A clear detection and attribution of cloud radiative effects to a perturbation in aerosol concentrations becomes possible when sub-filtering of the cloud field is applied, using the spatio-temporal distribution of the aerosol perturbation. Therefore, this work has implications for the detection and attribution of effective cloud radiative forcing in marine stratocumuli, which constitutes one of the major physical uncertainties within the climate system. Our results suggest that ships may sometimes have a substantial radiative effect on marine clouds and albedo, even when ship tracks are not readily visible.
Aerosol–cloud interactions (aci) in low-level clouds, which span just over a
fifth of the Earth's ocean surface
To reduce this uncertainty substantially through the use of satellite retrievals
and global climate models (GCMs) remains challenging. These challenges
include issues of collocation in retrievals of aerosol and cloud properties
from space
Databases obtained from satellite
retrievals
However, ship tracks are rare in comparison to the number of ocean-going
ships that criss-cross the world's oceans
The potential for albedo changes is particularly high in the open-cell and
disorganised stratocumulus regimes, which occur more frequently in the
subtropics, than in the closed-cell regime
Within this study we quantify changes in cloud radiative properties due to
aerosol perturbations in deep (boundary layer depth of
This study is based on a well-documented case of open-cell stratocumulus
clouds embedded within a boundary layer
Spatio-temporal averages of liquid water path (LWP), surface
precipitation (
The cloud regime was sampled during the early morning hours (03:00 to
08:30 local time) on 28 October 2008. A summary of cloud
properties measured during the campaign is given in Table
The open-cell clouds coincided with moister sub-cloud layer air masses, as
compared to the neighbouring closed-cell regime, and were characterised by
low sub-cloud layer aerosol concentrations (30
Two simulations were performed using the Weather Research and Forecasting (WRF) model
at the convection-resolving scale, with a horizontal grid resolution of
Both simulations were run for 48 h with a fixed divergence rate of
The simulations were performed with the two-moment
In addition to the control simulation, from here on named
In order to assess the radiative effect of concentrated and localised aerosol
pollution on deep open-cell clouds, the simulations need to demonstrate
sufficient skill in capturing the characteristics and dynamics of the
open-cell regime. Following initialisation, an unorganised stratiform cloud
deck formed in the ctrl simulation. Initial organised structures
appeared 6 h after initialisation following the onset of precipitation
(Fig.
The diurnal evolution of LWP and
The ctrl simulation was characterised by a well-mixed cloud layer
and stably stratified sub-cloud layer (see Fig. S1 in the Supplement), which is characteristic
of deep boundary layers. This structure developed rapidly following
initialisation from the well-mixed state. Within the first
Although simulated mean
The microphysical quantities, such as the mean sub-cloud layer aerosol
concentration (
In summary, despite remaining biases in the mean LWP and
Same as Table
The sea salt perturbed simulation displayed a spatially constrained aerosol
plume meandering around the emission line (Fig.
Inside the seeded region the emitted aerosol was predominantly transported
into the cloud within the updrafts of the cell walls (Fig. S3). Despite
efficient wet removal processes within the cell walls, the largest absolute
changes in
Snapshots of
The largest decreases in the cloud droplet effective radius at the cloud top
(
The changes in cloud-microphysical properties led to an increase in
domain-averaged LWP (Table
Due to the reduction in
Mean
Occurrence rate
The changes in domain-mean
The increase in cloud fraction alone, while assuming no further changes in
in-cloud
Figure
The increased occurrence of moderate LWP values (
Although the areal coverage of the detrained cloud amount between the cell
walls of the open cells increased, which contributed to the brightening of
the cloud deck, the highly concentrated aerosol perturbation was insufficient
to induce a transition from open to closed cells in these simulations.
Aerosols may impact this transition via aerosol–precipitation interactions.
Decreases in
While a ship track formed in the shallow boundary layer with open cells
between 10 and 15 km in size (Fig.
A change in
Knowing the position and extent of the aerosol perturbation allows one to
remove a sufficient amount of variability within the ship simulation
to obtain a spatially constrained, detectable and attributable response
within the cloud properties. As one averages along the spatial dimension of
the aerosol perturbation (coinciding with the
However, while changes in total albedo induced within the seeded region may
be identified in this manner, the change in total albedo of 0.03
(Table
Estimating the aerosol-induced radiative forcing in low-level marine clouds
constitutes a considerable uncertainty in the overall cloud radiative forcing
of anthropogenic aerosols. Satellite-based estimates of CRE changes due to
ship exhaust have remained inconclusive due to the high degree of variability
within the natural cloud scene
In this study we demonstrate that non-negligible amounts of brightening due
to anthropogenic shipping emissions may persist in the absence of a clear
ship track. In deep open cells, perturbations in
Furthermore, while these simulations are highly idealised in their set-up,
they do not necessarily reflect unrealistic emission conditions. The
prescribed ship is assumed to travel periodically along an identical emission
line without any crosswind, which may alter the plume size or dilute
emissions more effectively. Within the 48 h simulation, a total of five ships
traverse the
Increases in cloud-scene albedo were attributed to changes in brightness
within the stratified, detrained cloud regions covering the boundary layer
between convective cell walls. These detrained cloud regions are optically
thin (
In summary, our results suggest that although detectable ship tracks are
extremely rare in deep boundary layers, an increase in
Hovmoeller diagrams of
While these simulations are limited in their generality, they do demonstrate
that substantial changes in
Our results strongly motivate further research into the efficacy of aerosol
perturbations in deep open-cell stratocumulus. Here we demonstrate that the
aerosol forcing in this regime could be substantial. Yet for a clear
assessment the occurrence rate and magnitude of
The analysis of ship tracks and changes in cloud radiative properties within
them has arguably provided an extremely useful framework to develop a
mechanistic understanding of aerosol–cloud radiative interactions and to
constrain the effective cloud radiative forcing within marine low-level
clouds. However, linear shaped tracks are extremely rare and tend to form in
shallow boundary layers with a top below
800 m
At least 70 % of marine stratocumuli form in deeper boundary layers,
where distinct ship tracks due to ship emissions are very rarely detected.
Furthermore, 73 % of all stratocumuli globally are likely to occur within
the open-cell or disorganised regime Albedo changes equivalent to albedo increases in previously observed
ship tracks within shallow open-cell stratocumuli were embedded within a
stratocumulus deck of deep open cells, despite the absence of a spatially
coherent structure such as a ship track. The domain-mean all-sky albedo
increased by 0.05 due to a prescribed seeding source (sea salt emission
moving at 5 Regional changes in The simulated cloud brightening was attributed to the brightening of
the detrained cloud filaments that spanned the regions between the convective
cell walls of the open cells. These so-called veil clouds occur frequently in
low-level cloud layers and are connected to sub-cloud aerosol sources through
the convective cloud cores within the cell walls. Within these clouds the
brightening was largely attributed to increases in cloud fraction, with a
secondary contribution to brightening due to changes in cloud microphysical
properties.
The simulations for this paper are stored at
The supplement related to this article is available online at:
AP, HW, KC, RW and TPA designed the research. AP performed the research. HW contributed code adjustments to WRF. AP analyzed data. AP, HW, KC, RW and TPA wrote the paper.
The authors declare that they have no conflict of interest.
We acknowledge the Fund for Innovative Climate and Energy Research grant for the financial support of this research and the high-performance computing support from Yellowstone (ark:/85065/d7wd3xhc). This support was provided by NCAR's Computational and Information Systems Laboratory, sponsored by the National Science Foundation. Hailong Wang acknowledges support from the U.S. Department of Energy (DOE) Office of Science, Biological and Environmental Research. The Pacific Northwest National Laboratory (PNNL) is operated for DOE by Battelle Memorial Institute under contract DE-AC05-76RLO1830. Edited by: Patrick Chuang Reviewed by: two anonymous referees