Monitoring airmass/moisture boundaries with MSG

By Jochen Kerkmann (EUMETSAT) and Daniel Rosenfeld (HUJI)

MSG has two water vapour (WV6.2 and WV7.3), one ozone (IR9.7) and one carbon dioxide channel (IR13.4) to observe air mass characteristics such as moisture content, ozone content, tropopause height and static stability. In addition, MSG has two window channels (IR8.7 and IR12.0), which allow one to retrieve parameters such as total precipitable water and lower-level humidity.

Because water vapour absorbs slightly more at the IR12.0 than at IR10.8 window channel, the brightness temperature differences (BTD) IR10.8–-IR12.0 is directly related to the humidity content, which is dominated by the low-level humidity (assuming a normal (non inverted) temperature profile and neglecting surface emissivity effects).

This means, the larger the BTD (e.g. IR10.8–IR12.0) the higher the low-level absolute humidity. However, this only works under 'standard' atmospheric conditions when the temperature, especially in the lowest layers, decreases with height. In the case of low-level isothermal conditions or deep temperature inversions the BTD will be very small (or even negative) and not related to the humidity content.

Among the recommended RGB composites for MSG, the Dust RGB uses both BTDs mentioned above: the BTD IR12.0–IR10.8 on the red colour beam (range: from -4K to +2K) and the BTD IR10.8–IR8.7 on the green colour beam (range: from 0K to +15K). In addition, it uses the IR10.8 channel on the blue colour beam (range: from 261 to 289K). Thus, besides monitoring dust storms and tracking of thin, high-level clouds, it is also useful for the detection of low-level moisture boundaries.

During daytime (neglecting again spatial variability in surface emissivity), moist air will have a darker (blue) colour than dry air because of a more negative BTD IR12.0–IR10.8 (typically around -3K for moist air and -1 to 0K for dry air). The colour difference can be enhanced by narrowing the range for this BTD from [-4K, +2K] to [-3K, 0K], but at the cost of detectability of dust storms (the same applies to the range of the BTD IR10.8–IR8.7). During night-time, the detection of moisture boundaries is reduced because of low-level temperature inversions.

A good example of the monitoring of moisture boundaries is shown below. The initial situation at 01:00 UTC on 4 October 2005 (upper left image) shows a relatively moist airmass in the central part of the image (also marked by cloudiness) and a much drier airmass further to the west. The boundary is marked by a colour jump (from light blue to darker blue when going from the dryer to the more moist air masses) and by dust.

Later during the day, new convective development takes place along the moisture boundary and within the moist air. These new convective storms trigger new convective outflow boundaries that form a semi-circle around the storms. The outflow boundaries are marked by either dust storms and/or moisture boundaries, and sometimes also by lines of cumulus and Cb clouds (see. e.g. image at 05:00 UTC on 5 October 2005).

These new outflow boundaries may eventually intersect with old boundaries, like one can see on the lower left image. Those intersection points are likely areas for further convective development. Finally, on the last image (at 20:00 UTC on 5 October 2005) one can see that most of the Cbs have dissipated but that the outflow boundaries still exist to the east and to the west of the old storms.