Posts made by setvak

@TZ
You can try using some of the color palettes which are downloadable from the link above, or you can contact me directly by email. I would highly appreciate if the colors and their range can be the same (or very close) as on our website (http://portal.chmi.cz/files/portal/docs/meteo/sat/data_jsmsgview.html >> IR-BT), this way we would not confuse the public. EUMETSAT also uses the same color palette.
Re: (not) shifting the color range: this was rather a note that other weather services modify the range depending on their latitude or geographical region; for global images (as those used by you) this is impossible. BTW, the color scale usability also depends at the given location on actual height of the tropopause - with lower tropopause (e.g. polar air intrusions) the storm top temperatures are generally warmer, thus in those cases the storm tops appear in IR-BT images more blueish ...

@Gkikas-LGPZ

The reason for selecting 240K is that most storm anvils are at roughly this temperature (their warmest parts). When we were testing this color scheme about 12 years ago, we wanted to have the blue approx. outlining the storm anvils. Warmer temperatures depict also some of the lower, mid-level clouds, which we wanted to avoid.

@Gkikas-LGPZ

Detailed description of why we (weather satellite community) use this color scheme can be found e.g. here: https://cwg.eumetsat.int/color-enhancements/

In brief - the color layout follows:

1. Wien's displacement law - blue color represents warmer range, red coldest range (the same principle as color of stars in astronomy)
2. Red is for a human brain a signal of some form of danger - in case of convective storms the most dangerous parts of storm tops are the highest tops, which are also the coldest ones.

The typical color enhancement range for European (mid-latitude) storms is between 200K (red) and 240K (blue). For higher latitudes it needs to be shifted to the warmer range, while for tropics and subtropics it needs to be shifted to 190-230K, for the highest storms to 185-225K. It does NOT represent actual distribution and intensity of precipitation, though the coldest parts of storms are in most cases precipitating. In some cases even cirrus clouds (namely those in mountain waves or in a jet-stream) can be very cold, but without any precipitation.

Martin Setvák, CHMI