A sharp low level moisture boundary that stretched over the central Europe in June 2018, was well observed by both the SEVIRI Dust RGB product and MODIS 0.935 µm channel.
By Ivan Smiljanic (SCISYS)
A very dry cold front, with almost no clouds, was related to dry air advection from the north-east in central parts of Europe (Germany toward Austria and Balkans). The dew point temperatures dropped from around 15 °C to 0 °C, or less, in only a few hours. The standard SEVIRI Dust RGB product shows this vividly in the cloud-free areas, where the red component of this RGB product (Brightness Temperature Difference between 12.0µm and 10.8 µm channel) is mostly responsible for blue to pink gradients (Figure 1). The lighter blue and pink shades reveal less of the low-level water vapour content.
Figure 1: Meteosat-11 Dust RGB animation, 30 June 00:00 UTC–23:00 UTC. Download animation (MP4, 1MB)
A closer look at Central Europe reveals the dry areas in the region — the most pronounced moisture boundary stretching from north west to south east Germany, and other pronounced dry areas in central Belgium and east France (close to Lake Geneva).
Other areas with pink shades, with a rather granular texture, are not cloud-free areas, these correspond to relatively shallow cumulus fields in the observed domain (Figure 2).
Upper-level soundings from two stations in Germany (see sounding positions) confirm the presence of dry, i.e. moist low level air around the moisture boundary (Figures 3 and 4).
In Figure 5 the MODIS 0.935µm channel is compared with the SEVIRI channel difference 12.0–10.8 µm (enhanced and colour coded).
Light in the solar spectral wavelengths around 935 nm is typically absorbed by low level moisture, before or after reaching the ground. In that way darker areas of clear skies reveal more of water vapour content gradients in the lower levels of the atmosphere. A similar channel will be part of a 16 channel suite of the Flexible Combined Imager (FCI) on board the MTG-I geostationary satellite, at the moment it only exists on polar-orbiting instruments.
It appears that the SEVIRI BTD 12.0–10.8 µm difference is more sensitive to the low level moisture than 935 nm channel and, thus, shows more apparent blue gradients. However, this conclusion is biased on the fact that SEVIRI is looking at the scene at a relatively high angle, meaning that the moisture layer appears thicker than it really is (due to a longer optical path through the atmosphere). The 0.935 µm channel has better resolution (of nominally 1 km versus 3 km for SEVIRI channels) which leads to better definition of moisture boundaries. During the MTG era the difference in resolution between infrared channels (also used to build BTD products) and solar channels will nominally be 2 km v 1 km at the sub-satellite point.
Being a ‘solar channel’ 0.935 µm also reveals cloud contamination in the observed area and, to a degree, avoids the ambiguities that may appear with SEVIRI BTD difference. It is worth noting that the pale blue areas in the BTD 12.0–10.8 µm image do not appear as such with the 0.935 µm channel (for instance east area of France close to Strasbourg). This might be due to a difference in emissivity of different surfaces and not related to water vapour content — also a possible source of ambiguities with SEVIRI moisture detection.
The MODIS instrument has three dedicated channels in the spectral domain around 900 nm, namely 0.904 µm, 0.935 µm and 0.93 6µm (Figure 6). Visual comparison between three of those (at the same moment, with the same scanning and image rendering conditions) reveals that the best sensitivity to low level moisture is with 0.935µm channel. The 0.904 µm channel appears not to be sensitive to moisture content at all. FCI channel will be centred to around 0.91 µm.