Synoptic situation
The previous chapter already provided a brief introduction of the dust storm that overcasted Sardinia on July 28th, 2005. In this chapter we will study the event in more detail using several Meteosat 8 channels (starting with IR10.8) and channel combinations, along with some basic NWP. Clicking the image will bring you the appropiate loop that comes with the explanatory texts.|
Meteosat 8 - IR10.8: time sequence Can you detect the dust storm? The dust cloud originating from Algeria can be seen as it moves over the Mediterranean towards Sardinia. Once dust is scoured from the surface, it cools as it is lifted aloft. Thus, the dust cloud appears relatively cool in contrast to the relatively warm sea surface. Remember that this case happened during high summer and that the temperature of the Mediterranean was high resulting in a high temperature contrast with dust aloft! The cloud in this case reveals a strong signal in IR10.8 (inverted in the usual way) which makes it easy to dicriminate. However, as the temperature difference between dust and land or sea surface decreases, the IR10.8 window channel is little to no use in detecting dust. |
|
|
Meteosat 8 RGB - HRVis; HRVis; IR10.8: time sequence A second way to detect dust is by making use of the high resolution visible channel onboard Meteosat 8. In combination with IR10.8 an artificial RGB is made. In the previous set of images it was already seen that the IR contibution of the dust was high. In this RGB composite, the dust is best visible in the morning hours due to the strong forward scattering of the solar radiation on the dust particles, which are of roughly the same size as the wavelength of the incident solar radiation. During local noon, when sun and satellite are looking with nearly the same angle at the Dust cloud, the dust becomes almost invisible. You can click here to see a zoom of the region of interest. |
|
|
Meteosat 8 RGB - IR12.0-IR10.8; IR10.8-IR8.7; IR10.8: time sequence A third way of visualising the dust cloud is by making use of the Dust-RGB. This RGB was specifically designed to monitor the evolution of dust storms over mainly desert areas during both day and night. By making use of solely infrared channels this RGB composite makes it possible to track dust clouds 24 hours a day. The Dust-RGB is comprised of the following channels:
|
Above the Dust-RGB of 28 July at 0330UTC is presented. Immediately above it are the three individual composites of which the RGB is made from. You can shift and squeeze the bars and see what happens with the RGB image. By clicking "Probe" you can use your mouse to hover over the image and find out what each of the pixel values represent. The dust cloud, which is presented in pink, suggests a high value of red, some green and medium to high amount of blue. What the other colours mean in this RGB can be derived from the RGB-
|
| Clicking the image to the right will bring up the animation of the Dust-RGB. Remarkable is that already at the initial stage dust can be observed over the Mediterranean, stretching from the Baleares over Sardinia and Sicily. Some of this dust even moves further North towards the French Cote d’Azur that night. During the morning of the 28th a dust can be seen over Sardinia. In an anticyclonic curve it slowly moves over Sicily again into Northern Africa. |
|
So far the dust storm has been studied using various satellite images such as IR10.8 and RGB composites. This is adequate in nowcasting, but in order to forecast the movement of the dust cloud, numerical parameter fields must be combined with images. Useful basic parameters for dust clouds are the geopotential height and wind vectors at lower levels. To achieve this the ECMWF model is used.
|
Meteosat 8 - Geopotential Height 1000 hPa: time sequence The surface analysis of July 28th 2005 shows a high pressure system situated over Italy. Further south over northern Africa still a small part of the capital letter ‘T’ is seen which indicates a low pressure area near the surface. This latter system might be the cause of swirling the dust up high (enough) in the air. The loop to the right shows that there is no connection between the dust flow and isohypses. This was already expected after studying the signal of the dust in IR10.8 imagery, which was quite high. This led to the assumption that the dust cloud must be found higher aloft. |
|
|
Meteosat 8 - Geopotential Height 850 hPa: time sequence Alternatively we can also study the isohypses at 850 hPa. This is a level in which dust may normally be expected. After clicking the image and studying the various image in the timeline again unfortunately no coincidence between the isohypses (through the flow they indicate) and the dust flow is seen. |
|
|
Meteosat 8 - Geopotential Height 500 hPa: time sequence At 500 hPa the dust follows more the isohypses. The dust flows anticyclonically around an upper level high situated over Sahara. The high altitude of the dust also explains why the signal is strong in channel IR10.8: the dust has cooled significally while ascending. From this loop we can conclude that a good mesoscale model can give you a good insight of the stream directing the dust flow. Second, that the dust is found at this level also explains well why there is a quite strong signal found of the dust in IR10.8, e.g. it cooled significant as it moved aloft. |
|
|
Meteosat 8 - wind vectors 500 hPa: time sequence A second way of approaching the duststorm with numerical parameters is with the use of the wind vectors. After studying the previous set of images it is highly likely that the vectors at 500 hPa will coincide with the dust. To complete this analysis the wind vectors at 850 hPa are also presented. Studying the flow at 500 hPa the high over the Northern Sahara can easily be seen. This system directs the dust anticyclonically over the Mediterranean where it overcasts Sardinia and later parts of Sicily. |
|
Summary of investigations in this chapter
In this chapter the dustcloud that moved over Sardinia was studied using various satellite channels and basic numerical parameters. From these latter the geopotential height and the wind vectors prove to be the most usefull since thay can give a good indication in which direction the dust was moving. On the other hand, the effect of the dust on Sardinia and Sicily is not easy to conclude, because the dust was found high above the ground, at levels above the 850 hPa. Judging solely from satellite images it is difficult to estimate surface visibilities. Often blowing dust on the ground will appear as streaks parallel to the wind, while dust aloft in suspension will have a transparent, fuzzy appearance.
An indication for this was seen in the first set of images when the time sequence of Meteosat 8 IR10.8 images was analysed. A high contrasted picture discriminated the cold Dust clouds from the underlying warm Mediterranean. It was in this case that the dust could also be tracked in IR10.8, this is often not the case! After studying the basic parameters it was observed that the flow of the dust followed the geopotential height and wind vectors at 500 hPa. Giving some indication that the dust was found higher up in the troposphere. No exact figure however can be given from only NWP. A final indication on what level the dust is found, comes from the Cloud Type product developed by the Nowcasting SAF (NWCSAF). For this dustcase of July 28th over Sardinia the Cloud Type depicts low to mid-level cloudiness. The use of another product of the NWCSAF, namely the Dust flag will be introduced and discussed in the next chapter.