Better representation of cloud–aerosol interactions is crucial for an
improved understanding of natural and anthropogenic effects on climate.
Recent studies have shown that the overall aerosol effect on warm convective
clouds is non-monotonic. Here, we reduce the system's dimensions to its
center of gravity (COG), enabling distillation and simplification of the
overall trend and its temporal evolution. Within the COG framework, we show
that the aerosol effects are nicely reflected by the interplay of the
system's characteristic vertical velocities, namely the updraft (
Clouds are key players in the Earth's climate system via their influence on the energy balance (Baker and Peter, 2008; Trenberth et al., 2009) and hydrological cycle. Of all of the anthropogenic effects on climate, aerosol's effect on clouds remains one of the most uncertain (Boucher et al., 2013). In warm clouds, aerosol acts as cloud condensation nuclei (CCN) around which droplets can form, and therefore aerosol amount and properties determine the initial number of droplets and their size distribution (Squires, 1958; Rosenfeld and Lensky, 1998; Andreae et al., 2004; Koren et al., 2005). The initial droplet concentration affects cloud dynamics via microphysical and dynamical feedback throughout their lifetime. For example, the onset of significant collision events between droplets in polluted clouds, which are initially smaller and more numerous than in clean clouds (Squires, 1958), is delayed (Gunn and Phillips, 1957; Rosenfeld, 1999, 2000; Squires, 1958; Warner, 1968). This delay can have opposing effects on cloud development by increasing both the water loading (which reduces cloud buoyancy and vertical development) and the latent heat release resulting from the longer and more efficient condensation (increasing cloud buoyancy and vertical development) (Dagan et al., 2015a, b; Pinsky et al., 2013; Koren et al., 2014). We note that, often, these opposing effects act at different stages of the cloud's lifetime, further complicating the prediction of overall trends.
Air vertical velocities (
The system has another characteristic velocity that measures droplet
mobility. This velocity, defined as the effective terminal velocity (
Khain et al. (2005) used a two-dimensional cloud model with spectral (bin)
microphysics to study the aerosol effect on deep convective cloud dynamics.
They concluded that one of the reasons for comparatively low
Seigel (2014) showed an increase in
It has been recently shown (Dagan et al., 2015a, b, 2017) that, under given
environmental conditions, warm convective clouds have an optimal aerosol
concentration (
In this work, a bin-microphysics cloud model and large eddy simulation (LES)
of a cloud field were used to explore how changes in aerosol concentration
affect
The Tel Aviv University axisymmetric nonhydrostatic cloud model (TAU-CM) with detailed treatment of cloud microphysics (Reisin et al., 1996; Tzivion et al., 1994) was used. The included warm microphysical processes were nucleation of droplets, condensation and evaporation, collision–coalescence, breakup and sedimentation. The microphysical processes were formulated and solved using the method of moments (Tzivion et al., 1987).
The background aerosol size distribution used here represents a clean
maritime environment (Jaenicke, 1988). The aerosols are assumed to be
composed of NaCl. The different aerosol concentrations (25, 500 and
10 000 cm
The model resolution was set to 50 m, in both the vertical and horizontal
directions, and the time step to 1 s. The initial conditions were based on
theoretical atmospheric profiles that describe a tropical environment
(Malkus, 1958) (see profile T1RH2 in Fig. 1 in Dagan et al., 2015a). They
consisted of a well-mixed sub-cloud layer between 0 and 1000 m, a
conditionally unstable cloudy layer (6.5
To examine the effect of aerosols on the entire cloud, the properties
presented in this work are cloud mean values weighted by the liquid water
mass in each grid cell. Cloudy grid cells were defined as cells with liquid
water content larger than 0.01 g kg
The
To be consistent with the COG point of view, the mean air vertical (
We used the System for Atmospheric Modeling (SAM) LES model (Khairoutdinov
and Randall, 2003) with a bin-microphysics scheme (Khain and Pokrovsky, 2004)
to simulate the BOMEX (Barbados
Oceanographic and Meteorological EXperiment) warm cumulus case study (Holland
and Rasmusson, 1973; Siebesma et al., 2003). The horizontal resolution was
set to 100 m and the vertical resolution to 40 m. The domain size was
12.8
Cloud evolution on the phase space span by
Starting from the single-cloud scale, we first followed the entire cloud mean
We note that the vertical change in the COG height is determined by changes
in the vertical distribution of water mass due to microphysical processes
like condensation, evaporation and removal of mass by rain (in addition to
movement according to
Figure 1 demonstrates the importance of the aerosol effect on both
Shifting our view from the single-cloud scale to the cloud-field scale adds another layer of complexity as clouds affect the way in which the whole field's thermodynamics evolve with time. Moreover, 3-D simulations account for the effect of wind shear. Aerosol concentration has recently been shown to determine the trend of this evolution (Dagan et al., 2016, 2017). Clean precipitating clouds act to consume the initial instability that created them by warming the cloudy layer (in which there is net condensation) and cooling the sub-cloud layer (by rain evaporation). On the other hand, polluted non-precipitating clouds act to increase the field's instability by cooling and moistening the upper cloudy and inversion layers.
Temporal and spatial averages of the ambient air vertical velocity
(
Figure 3 presents the domain's mean
In the single-cloud-scale analysis (Sect. 3.1), we show how the timing of the
evolution of the two velocities dictates the aerosol effect. Here, having
many clouds in the field in different stages of their lifetimes, we first
analyzed the bulk properties of the two velocities. With the intention of
quantifying the relative contribution of the aerosol effect on the mean COG
height by modulating
To include the aerosol effect on the cloud-field thermodynamic properties, we
divided the simulation period into three equal thirds (excluding the first 2
h, each third of a period covered 4 h and 40 min). The
The decrease in the
In Fig. 3a, the presented quantities are domain and time averages. Figure 1
shows that the relative contribution of
Linear regression slope on the
As shown for the cloud scale, one of the most notable aerosol effects can be
viewed as delaying the onset of significant collection processes in the
polluted clouds (Koren et al., 2015) and therefore delaying the increase in
To quantify the evolution of the thermodynamic instability with time as a
function of aerosol loading (on a cloud-field scale), we looked at the time
trends in the
The sign of
Clouds form a complex system in which microphysical and dynamical processes are tightly linked and modulated by the thermodynamic properties of the environment. In turn, on the cloud-field scale, clouds affect the field's thermodynamic conditions. The aerosol effect on the droplet size distribution therefore affects all of the above. Better process-level understanding of the aerosol effect on cloud and rain properties in the case of warm convective clouds is essential for improving our understanding of the climate system. In this study, our aim was to better understand and quantify the aerosol effect on the air vertical velocity and droplet terminal velocity. Both characteristic vertical velocities' quantities modulate the distribution of water along the atmospheric column and hence affect the radiation (Koren et al., 2010) and heat balance (Khain et al., 2005). The findings presented here for the single-cloud and cloud-field scales could be used in future works to better represent cloud–aerosol interactions in coarser-resolution models (like climate models) as they provide a compact way to represent aerosol effect on the liquid water vertical mass flux and clouds' effect on the thermodynamic conditions.
Analyzing the two characteristic velocities on the cloud scale allows
separation, as a first approximation, between the aerosol effects on
condensation and evaporation efficiencies
(reflected by the magnitude of
Similar to the single-cloud case, the cloud-field (LES) results (that unlike
the single-cloud simulations account for 3-D processes such as wind shear)
demonstrated an increase in
Using a cloud-tracking algorithm, we identified the growing stage of the
clouds and examined the relative contribution of the aerosol effect on COG
height by modulating
No data sets were used in this article.
The authors declare that they have no conflict of interest.
This research was supported by the Minerva Foundation with funding from the Federal German Ministry of Education and Research and the De Botton Center for Marine Science. Maria Cristina Facchini Reviewed by: two anonymous referees