EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Satellite Observations of Organizational Regimes in Low-Level Mixed-Phase Clouds over the Southern Ocean

Jessica Danker1, Odran Sourdeval2, Isabel L. McCoy3, Robert Wood3, and Anna Possner1
Jessica Danker et al.
  • 1Institute for Atmospheric and Environmental Sciences, Goethe University Frankfurt, Frankfurt, Germany (Contact:
  • 2Laboratoire d’Optique Atmosphérique, Université de Lille, Villeneuve-d’Ascq, France
  • 3Atmospheric Sciences, University of Washington, Seattle, WA, USA

On average stratocumulus clouds cover about 23% of the ocean surface and are important for Earth’s radiative balance. They typically self-organize into cellular patterns and thus are often referred to as mesoscale-cellular convective (MCC) cloud systems. In the Southern Ocean (SO), low-level clouds cover between 20% to 40% of the ocean surface in the mid-latitudes where they exert a substantial radiative cooling. In a previous study, McCoy et al (2017) demonstrated that different MCC regimes may be associated with different cloud albedos and thus different cloud radiative forcing.
Many of the MCC clouds in the SO are not pure liquid but contain a mixture of liquid and ice. Here we investigate whether the formation of ice within these mixed-phase clouds influences MCC organization and thus the cloud-radiative effect.
To investigate the cloud phase we use the raDAR-liDAR (DARDAR) data product (version 1) from Cloud-Aerosol-Water-Radiation Interactions (ICARE) Data and Services Center which provides collocated data from Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO), CloudSat and Moderate Resolution Imaging Spectroradiometer (MODIS). The “Simplified DARMASK Categorization Flag” of DARDAR is used to categorize the vertically resolved cloud phase into a single cloud phase per data point: clear, multi-layer, liquid, mixed or ice. In order to distinguish between open and
closed MCC regimes, we collocate the DARDAR product with an MCC classification data set from McCoy et al (2017) which is based on a neural network algorithm applied to MODIS Aqua data.
Our preliminary results confirm previous ground-based observations that most mixed-phase clouds are composed of a supercooled liquid top and ice underneath. Furthermore, our preliminary analysis suggests open MCCs occur more frequently as mixed-phase clouds (57% (DJF), 55% (JJA)) in the SO compared to liquid clouds (39% (DJF), 37% (JJA)) during both summer (DJF) and winter (JJA). In contrast, closed MCCs are more likely to appear as liquid clouds (58%) in comparison to mixed-phase clouds (40%) during winter, whereas during summer there seems to be no tendency for closed MCCs to be either liquid (51%) or mixed (49%).

How to cite: Danker, J., Sourdeval, O., McCoy, I. L., Wood, R., and Possner, A.: Satellite Observations of Organizational Regimes in Low-Level Mixed-Phase Clouds over the Southern Ocean, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-13229,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 18 Apr 2020
  • CC1: Comment on EGU2020-13229, Andrew Gettelman, 04 May 2020

    Hi Jessica,

    Nice analysis. Does the size/structure (Open/Closed) of the cloud influence detection capability for DARDAR? I.e. would observational sampling bias the results?  

    • AC1: Reply to CC1, Jessica Danker, 05 May 2020

      Hi Andrew,

      we collocated the data set from McCoy et al. (2017), which is based on MODIS, with the DARDAR data set to identify the MCC regimes. So DARDAR itself is not identifying the MCC structures. Then we tested if the overall occurrence frequency of closed and open MCCs in DARDAR-only regions in JJA and DJF is comparable to the occurrence frequency in McCoy et al (2017) of the Southern Hemisphere (30 to 60°S slightly different from our definition of SO) and found similar ratios of open and closed MCCs for either season. Therefore, it seems that the cloud structure is not influencing the detection capability of DARDAR. It is similarly likely to retrieve in open and closed cells.

      Concerning the capability of DARDAR to detect the cloud phase in both regimes, the DARDAR-MASK algorithm is determining the cloud phase solely based on the CPR and the temperature when the lidar signal extincts which could bias the detection of cloud phase when one of the regimes would have a higher vertical extent. However, we also tested this, while open MCCs have a slightly higher CTH (140m in DJF, 280m in JJA) than closed MCCs in 2008, they also show a slightly higher CBH (140m in DJF, 240m in JJA) than closed MCC. This would suggest that the cloud phase identification between both regimes should not be biased to one regime.