Cellular convection in cirrus clouds as a possible effect of dust aerosols

Cellular convection in cirrus clouds

11 February 2010 00:00 UTC

Cellular convection in cirrus clouds as a possible effect of dust aerosols
Cellular convection in cirrus clouds as a possible effect of dust aerosols

Cellular convection in cirrus clouds as a possible effect of dust aerosols.

Last Updated

22 October 2020

Published on

10 February 2010

by Kornel Kollath (Hungarian Meteorological Service )

In rare cases cellular convective elements can appear at the top of high level cirrus clouds, which can be very similar in appearance to well-known low level closed-cells (Benard cells) over oceans (see Report ). A typical example of such a cellular cirrus shield was found over Eastern Europe on 10–12 February 2010 (see image below). The cloud shield developed on the forward side of a Mediterranean cyclone and could be analysed as warm-conveyor belt or warm frontal shield (see Report, Figure 4).

During the development the shield was situated around the axis of an upper level ridge (see Report, Figure 5). In the first part of its development extremely cold cloud top temperatures (-75 to -80 °C) appeared very quickly (see Animation , AVI) (with continuous cooling, especially between 10 February, 06:00 UTC and 11 February, 06:00 UTC). The structure of the cloud top is particularly noticeable in high-resolution NOAA images in the morning of 11 February (see NOAA images below).

According to preliminary research of MSG images in central and southern Europe, this cellular structure normally only develops in the situation of a southerly airstream from North Africa (Nagy, 2009). In many of these cases the history of the cloud shield can be associated with a detectable dust plume from Africa. In this situation dust can also be detected in the MSG images (see Dust RGB Composite , 10 Feb 2010 12:00 UTC).

The trajectories show that the Saharan dust aerosols could ascend up to cirrus levels with the upward motion that was present (see Backward Trajectory , source: ECMWF). We can also follow the route of the increased aerosol content by special forecasts for aerosol optical thickness. There was good correspondence between the position of the cirrus shield and the higher aerosol concentration in this case (see Report, Figure 10, and animation , AVIsource: University of Athens).

Increased aerosol concentration can change the microphysical properties and radiation processes of clouds. The development of cellular convection needs to have unstable stratification inside the cloud. An increased long wave radiation flux from the cloud top could be a possible hypothesis which can enhance the instability in the upper layer of the cirrus clouds (Nagy, 2009, see Conceptual Model ). Usually the cumuliform structure at the top can be very well seen in the morning, but during daytime this structure diminishes presumably because of the absorption of the solar radiation (see Conceptual Model ).


Nagy, Andrea, 2009: Investigating weather situations which bring Saharan dust over Hungary based on MSG satellite images. Master's thesis, ELTE University, Budapest (available in Hungarian language).

Note: On 11 February 2010, the Turkish State Meteorological Service (TSMS) reported coloured (dust) rain over Istanbul.


Meteosat-9 Image

Met-9, 11 February 2010, 06:45 UTC
RGB Composite HRV, HRV, IR10.8
Large Area

Related Content

Met-9 IR10.8/WV6.2 animation (9 Feb 17:30 UTC–11 Feb 07:00 UTC, AVI)
Met-9 IR10.8/WV6.2 animation (7 Feb 23:45 UTC–12 Feb 17:45 UTC, AVI)
Met-9 IR10.8/WV6.2 animation (with GFS 300 hPa winds and height of 2 PVU, AVI)
Met-9 IR10.8/WV6.2 animation (with GFS 300 hPa streamlines and height of 2 PVU, AVI)
Forecasting coloured rain in Bulgaria (23–24 March 2008)
Large dust swirl over Algeria (20–22 February 2007)
Dust storm over Northern Africa and the Eastern Mediterranean Sea (22–24 February 2006)
Dust storm over the Mediterranean Sea (21 February 2004)
Low-level Gravity Waves within a Layer of Bénard Cells as seen by Meteosat-7



NOAA-17, 11 February 2010, 07:53 UTC
NOAA-17, 11 February 2010, 07:53 UTC
Channel 04 (IR11.0)
Full Resolution
NOAA-17, 11 February 2010, 07:53 UTC
NOAA-17, 11 February 2010, 07:53 UTC
Channel 04 (IR11.0, colour enhanced)
Full Resolution

Additional Information (from Mike Fromm, NRL)

The cellular structure of the 'dusty' cirrus clouds can also be seen in the radar (Cloudsat) and lidar (CALIPSO) data (see image below). The track goes from Central Turkey (40 deg latitude) to Southern Finland (60 deg latitude). Unusually small radar reflectivities are observed for such an optically thick cloud. Note that parts of the thick cirrus cloud are completely transparent at the CloudSat frequency, which indicates the presence of very small ice particles. Unpolluted high clouds have similar tops in both lidar and radar data. Also, most of the dusty cirrus shield has no precipitation below.

A-train view of the 'dusty' cirrus cloud

A-Train view of the 'dusty' cirrus cloud shield on 11 February 2010 with CALIPSO 532 nm attenuated backscatter coefficient and CloudSat radar reflectivity plotted together (Cloudsat plotted over CALIPSO). Also the MODIS IR radiance, scaled between curtain min and max, is shown.
Full Resolution (source: Mike Fromm (NRL))