Hemispheric contrasts in ice formation in stratiform mixed-phase
clouds: Disentangling the role of aerosol and dynamics with
ground-based remote sensing
Martin Radenz
1
, Johannes Bühl
1
, Patric Seifert
1
, Holger Baars
1
, Ronny Engelmann
1
, Boris Barja
González
2
, Rodanthi-Elisabeth Mamouri
3,4
, Félix Zamorano
2
, and Albert Ansmann
1
1
Leibniz Institute for Tropospheric Research (TROPOS), Leipzig, Germany
2
Atmospheric Research Laboratory, University of Magallanes, Punta Arenas, Chile
3
Department of Civil Engineering and Geomatics, Cyprus University of Technology of Technology, Limassol, Cyprus
4
ERATOSTHENES Centre of Excellence, Limassol, Cyprus
Correspondence: Martin Radenz (radenz@tropos.de)
Abstract.
Multi-year ground-based remote-sensing datasets acquired with the Leipzig Aerosol and Cloud Remote Observations System
(LACROS) at three sites: a highly polluted central European site (Leipzig, Germany), a polluted and strongly dust-influenced
eastern Mediterranean site (Limassol, Cyprus), and a clean marine site in the southern mid-latitudes (Punta Arenas, Chile) are
used to contrast ice formation in shallow stratiform liquid clouds. These unique, long-term datasets at key sites of aerosol-5
cloud interaction provide a deeper insight into cloud microphysics. The influence of temperature, aerosol load, boundary-layer
coupling and gravity wave motion on ice formation is investigated. With respect to previous studies of regional contrasts in the
properties of mixed-phase clouds our study contributes the following new aspects: (1) Sampling aerosol optical parameters as a
function of temperature, the average backscatter coefficient at supercooled temperatures is within a factor of 3 at all three sites.
(2) Ice formation was found to be more frequent for cloud layers with cloud top temperatures above −15
◦
C than indicated10
by prior lidar-only studies at all sites. A virtual lidar-detection threshold of IWC needs to be considered in order to bring
radar-lidar-based studies in agreement with lidar-only studies. (3) At similar temperatures, cloud layers which are coupled to
the aerosol-laden boundary layer show more intense ice formation than de-coupled clouds. (4) Liquid layers formed by gravity
waves were found to bias the phase occurrence statistics below −15
◦
C. By applying a novel gravity wave detection approach
using vertical velocity observations within the liquid-dominated cloud top, wave clouds can be classified and excluded from the15
statistics. After considering boundary layer and gravity-wave influences, Punta Arenas shows lower fractions of ice containing
clouds by 0.1 to 0.4 absolute difference at temperatures between −24 and −8
◦
C. These differences are potentially caused by
the contrast in the INP reservoir between the different sites.
Copyright statement. TEXT
1
https://doi.org/10.5194/acp-2021-360
Preprint. Discussion started: 17 May 2021
c
Author(s) 2021. CC BY 4.0 License.
1 Introduction20
Clouds and aerosol are inseparably coupled, linked via complex pathways of interaction whose outcome manifests in the
macroscopic properties of precipitation and radiation fields. On the one hand, aerosol particles are required as cloud con-
densation nuclei on which cloud droplets can form. On the other hand, primary ice formation in the heterogeneous freezing
temperature range from 0 to approximately -40°C requires ice nucleating particles (INP) to be present in the aerosol reservoir.
The ways in which aerosol and cloud particles interact are controlled by the dynamics and thermodynamics of the atmospheric25
environment. Thermodynamic processes are considered to dominate the cloud microphysical properties because they control
the amount of water vapour that is available for being transferred to either the liquid or the ice phase. This dominance makes it
difficult to isolate aerosol-related effects in observations of cloud properties.
Nevertheless, observations as well as aerosol-permitting model studies suggest a considerable influence of the aerosol con-
ditions on the properties and evolution of clouds and precipitation (Seifert et al., 2012; Possner et al., 2017; Solomon et al.,30
2018; Zhang et al., 2018). Solomon et al. (2018) used high-resolution modelling of Arctic mixed-phase clouds to show, that
perturbations in the INP concentration dominate over changes in the cloud condensation nuclei (CCN) concentrations. Cloud
chamber studies suggest, that holding CCN constant, the ratio of ice to liquid water content in the steady state is predominantly
controlled by INP concentrations (Desai et al., 2019).
There is also a distinct spatio-temporal variability of the performance of weather and climate model simulations, which is35
attributed to the insufficient representation of aerosol-cloud-dynamics interaction processes in the models (Fan et al., 2016; Se-
infeld et al., 2016). For instance, the reasons for the less accurate treatment of the radiative balance in the southern-hemispheric
mid-latitudes compared to their northern-hemispheric counterpart are still debated (Trenberth and Fasullo, 2010; Grise et al.,
2015). The atmosphere of the southern hemisphere’s mid-latitudes is a unique component of the Earth’s climate system. It’s
the stormiest (e.g. Young, 1999) and one of the cloudiest places on Earth (Haynes et al., 2011; Naud et al., 2014), but process40
understanding of clouds in that region is still limited. The reported biases in the solar radiation budget are attributed to shallow
supercooled liquid topped clouds, which are insufficiently represented by current models (Bodas-Salcedo et al., 2014; Kay
et al., 2016; Bodas-Salcedo et al., 2016; Kuma et al., 2020). These radiation biases affect estimates of sea surface temperature
and surface precipitation, ultimately affecting the energy balance at the surface (Franklin et al., 2013; Hyder et al., 2018). There
is an ongoing controversy about the reasons for the observed differences and prevailing model deficiencies, but indications are45
given that a combination of hemispheric contrasts in aerosol load and atmospheric dynamics plays a role.
Different causes for the excess of the supercooled liquid cloud layers are proposed. On the one hand, reason for the excess
of supercooled liquid water in southern-hemispheric cloud systems could be the inhibition of ice formation caused by the
lack of ice nucleating particles in the predominantly pristine environment of the Southern Ocean (Hamilton et al., 2014), where
terrestrial sources, which are frequently considered a good source for INP, are rare or far apart (Vergara-Temprado et al., 2017).50
In the heterogeneous freezing regime, suitable aerosol particles are a prerequisite for ice formation and missing ice formation
as a sink for cloud water, the liquid phase may be sustained for long periods of time. On the other hand, dynamical processes
could lead to an enhancement of supercooled liquid water. Korolev (2007) showed, that depending on the number and size of
2
https://doi.org/10.5194/acp-2021-360
Preprint. Discussion started: 17 May 2021
c
Author(s) 2021. CC BY 4.0 License.
ice crystals, a threshold vertical velocity can be found, which allows for sufficient supersaturation to grow the ice as well as the
liquid phase. Gravity waves have been suggested to play a role in the phase partitioning of Southern Ocean clouds (Alexander55
et al., 2017; Silber et al., 2020). Due to the orographic effects and the strong westerlies, gravity waves are a general feature in
the vicinity of all landmasses in the middle and high latitudes of the southern hemisphere (Sato et al., 2012; Alexander et al.,
2016). It is however also noteworthy that stronger turbulence increases the amount of ice formed in stratiform cloud layers
(Bühl et al., 2019). Indications are thus given that it is necessary to also consider turbulence in studies of ice formation.
The undetermined contributions of dynamical and aerosol effects on the observed excess of supercooled liquid in the south-60
ern hemisphere requires dedicated attribution studies. In numerous previous studies, liquid-topped supercooled stratiform cloud
layers have been proven to be suitable natural laboratories for the investigation of the relationships between aerosol properties,
thermodynamics and microphysical properties of clouds in the heterogeneous freezing regime. Temperature at which the ice
formation occurs, needs to be strongly constrained, because the concentration of efficient INP increases rapidly with decreasing
temperature (e.g. Kanji et al., 2017). Accordingly, the amount of ice formed also increases for lower temperatures (Bühl et al.,65
2016). The lowest temperature in such clouds occurs on top of the liquid-dominated layer, hence the cloud top temperature
(CTT) can be used to constrain the ice formation temperature. Turbulence is usually confined to the liquid-dominated cloud top
(Westbrook and Illingworth, 2013; de Boer et al., 2009) and due to the limited thickness of this layer, secondary ice formation
or ice multiplication are strongly constrained (Fukuta and Takahashi, 1999; Myagkov et al., 2016). Contrasting microphysical
properties observed in pristine, clean regions with observations from areas with higher aerosol load, e.g., allows to advance70
understanding of the impact of different aerosol loads (Choi et al., 2010; Kanitz et al., 2011; Seifert et al., 2015; Tan et al.,
2014). Recent studies based on the A-Train satellite constellation suggest systematically lower ice amounts in the southern
mid-latitudes (Zhang et al., 2018) and a strong susceptibility to dust load (Villanueva et al., 2020). Supercooled liquid clouds
are frequent over the Southern Ocean (Huang et al., 2015; Hu et al., 2010) and studies by Kanitz et al. (2011) and Choi et al.
(2010) showed, that - at similar temperatures - ice is formed less frequently by liquid layers in the southern hemisphere mid-75
latitudes, than in the northern hemisphere. The study of Kanitz et al. (2011) first used a ground-based lidar at Punta Arenas
(53.1°S 70.9°W, Chile) to asses the thermodynamic phase of stratiform mixed phase clouds above the Southern Hemisphere
mid-latitudes. Major caveats of this study are the limited duration of the observations during austral summer and the limitations
of the lidar-only setup. Alexander and Protat (2018), using ground-based lidar and A-Train from Cape Grim (40.7°S 144.7°E,
Australia) confirm the basic findings also for the eastern parts of the Southern Ocean. This study, similarly to McErlich et al.80
(2021), emphasizes the problems in A-Train derived datasets detecting shallow clouds in the lowermost part of the atmosphere.
Recent activities include shipborne, land-based and aircraft campaigns targeting aerosols and clouds above the Southern
Ocean between Australia and Antarctica (60 to 160
◦
E). An overview is provided by McFarquhar et al. (2020). In terms of
stratiform clouds, the large abundance of supercooled liquid water, occasionally down to −30
◦
C, was confirmed. But, apart
the year-long lidar/radar dataset at Macquarie Island (54.6°S 158.9°E, Australia) the cloud observations focused on austral85
summer.
Using a shiporne dataset, Mace and Protat (2018) also found frequent liquid-dominated clouds with low radar reflectivities
and 1/3 of the liquid layers only observed with lidar. Comparing the observations with a Cloud-Aerosol Lidar and Infrared
3
https://doi.org/10.5194/acp-2021-360
Preprint. Discussion started: 17 May 2021
c
Author(s) 2021. CC BY 4.0 License.
Pathfinder Satellite Observations (CALIPSO) dataset from Hu et al. (2010), they found an overestimation of supercooled
liquid in the satellite dataset, especially strong at temperatures above −15
◦
C. In a follow-up study, Mace et al. (2020) refined90
the CALIPSO classification scheme, leading to more frequent detections of the mixed phase, especially during wintertime and
in the lower latitudes of the Southern Ocean. However, no CTT-resolved phase occurrence statistics is presented. Liquid layers
in deeper clouds, observed during another shipborne campaign (McFarquhar et al., 2020; Alexander et al., 2021), could only be
reproduced in regional model simulations, when INP parametrization was tuned to lower concentrations (Vignon et al., 2021).
Zaremba et al. (2020) investigated airborne active remote sensing observations of Southern Ocean clouds south of Tasmania.95
They also found widespread liquid cloud tops at temperatures down to −30
◦
C. By investigating the ground-based remote
sensing dataset assembled at McMurdo (77.8°S 166.7°E, Antarctica), Silber et al. (2018) found frequent long-lived liquid
topped clouds, also below −30
◦
C.
Yet, a statistical analysis of the relationship between both, aerosol conditions, cloud vertical dynamics, and the phase par-
titioning in stratiform cloud layers of the southern-hemisphere mid-latitudes based on long-term observations was not estab-100
lished. One reason is, that, despite increased activity in the recent past, ground-based remote sensing observations of clouds
and aerosol are still sparsely distributed in the Southern Ocean and at the coast of Antarctica.
Goal of this study is to analyze long-term ground-based remote sensing observations of aerosol properties, cloud micro-
physics and atmospheric dynamics from three sites with strongly contrasting aerosol conditions in order to attribute ice forma-
tion in the heterogeneous freezing regime to atmospheric dynamics and aerosol conditions. For that attribution approach, we105
utilized the recent campaigns of the Leipzig Aerosol and Cloud Remote Observations System (LACROS) at Leipzig (51.4°N
12.4°E, Germany), Limassol (34.7°N 33.0°E, Cyprus) and Punta Arenas (53.1°S 70.9°W, Chile) which provide datasets, that
cover the aerosol conditions of a continental northern hemispheric background site, a hot-spot of mineral dust, and the marine-
dominated pristine Southern Ocean, respectively. Hence, these datasets collected with a single set of ground-based remote
sensing instrumentation provide an ideal basis for contrasting studies. The broad variety of instruments covers the decisive110
properties of aerosols, dynamics, clouds and precipitation for a more comprehensive picture of aerosol-cloud interaction. The
observations at Punta Arenas provide the first multi-year dataset of synergistic ground-based remote sensing observations in
the western half of the Southern Ocean and allow to contextualize prior findings.
The paper is structured as follows: In section 2 the instrumentation and campaigns are described (Sec. 2.1), followed by
the retrievals of aerosol properties (Sec. 2.2) and the synergystic Cloudnet retrieval (Sec. 2.3). Afterwards, the methods for115
cloud selection and vertical velocity characterization are introduced. Instruments, field campaigns and synergistic retrievals are
described first. Section 2 covers the methods, consisting of the retrieval of the aerosol statistics, the automated cloud selection
algorithm (Sec. 2.4) and the characterization of vertical velocity (Sec. 2.5). The statistics is presented in Sec. 3, including the
lidar-derived average profiles of aerosol optical properties (Sec. 3.1) and the cloud phase statistics with special emphasis on
instrument sensitivity (Sec. 3.2.1). Boundary layer coupling (Sec. 3.2.2) and vertical dynamics are discussed in Sec. 3.2.3.120
The amount and efficiency of ice production is assessed in Sec. 3.2.4. The study concludes with a discussion of the contrast
identified in aerosol load and stratiform cloud properties (Sec. 4) followed by a summary and outlook (Sec. 5).
4
https://doi.org/10.5194/acp-2021-360
Preprint. Discussion started: 17 May 2021
c
Author(s) 2021. CC BY 4.0 License.
Table 1. Specifications of the LACROS instruments used in this study.
Instrument
(Reference)
Frequency ν
Wavelength λ
Measured quantity Temporal
resolution
Vertical range Vertical
resolution
Doppler cloud radar
METEK Mira-35
(Görsdorf et al., 2015)
ν = 35 GHz
Radar reflectivity factor 3.5 s 150 − 13000 m 30 m
Vertical velocity 3.5 s 150 − 13000 m 30 m
Linear depolarization ratio 3.5 s 150 − 13000 m 30 m
Raman-Polarization Lidar
Polly
XT
(Engelmann et al., 2016)
λ = 355, 532, 1064 nm Attenuated backscatter coeff. 30 s 100 − 15000 m 7.5 m
λ = 355, 532 nm Raman backscatter signal 1 h 300 − 5000 m ∼ 50 m
λ = 355, 532 nm Linear depolarization ratio 30 s 100 − 15000 m 7.5 m
Microwave radiometer
RPG HATPRO-G2
(Rose et al., 2005)
ν = 22.24 − 31.4 GHz Brightness temperatures 1 s column integral
ν = 51.0 − 58.0 GHz Brightness temperatures 1 s column integral
Doppler Lidar
Halo Streamline
(Pearson et al., 2009)
λ = 1.5 µm
Attenuated backscatter coeff. 2 s 48 − 12000 m 48 m
Vertical velocity 2 s 48 − 12000 m 48 m
Ceilometer
Jenoptik chm15kx
λ = 1064 nm Attenuated backscatter coeff. 30 s 15 − 15300 m 15 m
Optical disdrometer
Ott Parsivel
2
(Löffler-Mang and Joss, 2000)
λ = 650 nm Hydrometeor size distribution 30s 4 m -
Sun photometer
Cimel CE318-T
(Barreto et al., 2016)
λ = 340 − 1064 nm Aerosol optical thickness variable column integral
2 Data and Methods
This section introduces the datasets and methods used in the remainder of this study. Starting with the campaigns and in-
strumentation (Sec. 2.1), followed by a short description of the retrievals used (Sec. 2.2 and 2.3) and finally the methods for125
automated selection of shallow stratiform clouds (Sec. 2.4) and characterization of vertical velocity dynamics (Sec. 2.5).
2.1 Datasets from Leipzig, Limassol and Punta Arenas
Basis of the observational datasets presented below is the Leipzig Aerosol and Cloud Remote Observations System (LACROS),
the mobile ground-based remote-sensing supersite of the Leibniz Institute for Tropospheric Research (TROPOS), Leipzig,
Germany. The instrumentation used for the synergistic approaches applied in this study comprises a Mira-35 35GHz scanning130
cloud radar, a Polly
XT
multi-wavelength Raman and depolarization lidar, a Streamline 1.5µm scanning Doppler lidar, a HAT-
PRO 14-channel microwave radiometer, a 1064 nm ceilometer, an optical disdrometer and radiation sensors. Main properties
of the sensors are summarized in Table 1.
LACROS was established in 2011 and was first introduced by Bühl et al. (2013b). After the initial setup phase, the basic set
of instrumentation of LACROS has been kept unchanged since the year 2014. Hence, this comparative study uses data from the135
5
https://doi.org/10.5194/acp-2021-360
Preprint. Discussion started: 17 May 2021
c
Author(s) 2021. CC BY 4.0 License.
