Table of Contents
Introduction
Combining satellite images from different channels goes far back in the history of satellite meteorology. The idea behind it can be summarized as follows:
Different channels give different information, and the combinations of channels reveal more information and features of the Earth's surface and atmosphere than single channels alone.
There are two ways of making better use of satellite images:
- Enhanced satellite images
- RGB presentation (Red Green Blue)
Looking back at the history of satellite meteorology, the first enhanced images were generated for severe convection like MCS (Mesoscale Convective Systems) to reveal the coldest cloud tops and the growing rate of these cold tops. The first RGB combinations were developed for the USA's polar orbiting satellites and the first application was discriminating low clouds from middle and high clouds.
As time passed, a large number of different RGB products were developed by various institutes and weather services. The usefulness of some combinations was sometimes questionable and comparison between different RGBs became difficult.
With the creation of the MSG and its 12 channels, each with its own unique qualities, a series of standardized RGBs were developed. These RGBs will be described according to their significance and usefulness.
Basic ideas behind enhanced satellite images
This application typically uses IR images that represent cloud top temperatures. Different colors are assigned to specific temperature ranges.
The following example presents the same case as the one that was already used for the basic channels - a winter case. The coldest temperatures can be found in the cloud tops of a warm front shield; the high cloud fibers at the rear of the cold frontal cloud band are also enhanced by the corresponding colors.
However, enhanced IR images are especially useful for convective cloud systems; a summertime example with an unstable situation is shown in the case below. Beyond simply coloring the coldest clouds, enhanced IR images reveal structures typical for convective cells and systems like cold rings, warm cores, U/V structures and overshooting tops much more easily than basic channels alone would.
Basic idea behind RGBs
The three colors red, green and blue are allocated for three MSG channels or, if appropriate, channel differences. All other colors are generated as a combination of these three basic colors. The choice between channels and channel differences depends on the features that are being looked for.
This shall be demonstrated with the so-called Natural Color RGB, which is composed from the channels 03 (R), 02 (G) and 01 (B). Each of these three visible channels contains information about reflected sunlight (and consequently the optical thickness of clouds), but each channel also adds its own physical specialty:
Ch 03 (1.6 µm): particle phase and size (water, small and bigger ice droplets)
Ch 02 (0.8 µm): "greenness" of vegetation
Ch 01 (0.6 µm): optical thickness
The next figure shows the resulting colors for four typical features in the "Natural Color RGB"
Maximum signal from Ch 2: cloud-free; surface and vegetation dominates; resulting color: green | |
Medium values from all channels but Ch 3 dominates; no clouds or vegetation; resulting color: reddish | |
Medium values from all three channels; medium albedo - thick water cloud; resulting color: grayish | |
High values from ch 1 and 2 : high albedo; low counts from Ch 3: typical for ice cloud and/or snow; resulting color: cyan |
When using RGBs, it is important to understand which qualities the individual channels or channel differences provide.
The Natural Color RGB
Information received from the three channels that contribute to the RGB combination
NIR IR | Optical thickness and particle phase and size - information about water droplets as well as small and large ice droplets |
VIS 02 | Optical thickness and differences in vegetation - "greenness" of vegetation |
VIS 01 | Optical thickness: albedo - information about cloud thickness |
Resulting colors for typical features
Gray | Thick water clouds |
Cyan | Ice clouds, snow |
Green | Vegetation |
Red | vegetation free land, desert |
Black | Ocean |
Typical application areas
- First impression of the large scale weather
- Differentiation of water and ice clouds
The High Resolution VIS (HRVis) RGB
Information received from the three channels that contribute to the RGB combination
HRV | Optical thickness - High resolution (pixel: 1 km at Nadir) |
HRV | Optical thickness - High resolution (pixel: 1 km at Nadir)
The HRV channel is used in both red and green components to retain the high resolution information |
10.8 | (Classical IR Window channel) - Information about cloud top temperature |
Resulting colors for typical features
White | Thick convective cells |
Blue | Cold cloud tops, especially thin cirrus |
Yellow | Fog and low cloud |
Typical application areas
- Discrimination of cloud systems of different vertical extent
- Recognition of small-scale structures, especially in convective and low clouds/fog
This RGB is very useful for convective systems, as it lets us tell the thick Cb kernel apart from the thin high Cirrus anvil, as well as recognizing the overshooting tops. A convective summer case is presented below:
The Airmass RGB
Information received from the three channels that contribute to the RGB combination
WV6.2 - WV 7.3 (difference of the two WV channels) | Information about moisture content in middle/upper layers of the troposphere |
IR 9.7 - IR 10.8 (ozone channel - IR window channel) | Information about ozone concentration, which differentiates cold (polar) and warm (mid-level, subtropical) air masses
Cold air mass: high ozone concentration and low tropopause; warm air mass: low ozone concentration and high tropopause |
WV 6.2 (Water Vapor channel 05) | Information about humidity in upper layers of the troposphere |
Resulting colors for typical features
White | Thick clouds |
Green | Warm air masses |
Blue | Cold air masses |
Dark red/brown | Dry upper air, PV anomalies |
Typical application areas
- Differentiation between cold and warm air masses (blue - green)
- Recognition of very dry air that has descended from the stratosphere (dark brown)
- Dynamical features: tropopause folding; high PV values
Due to the incorporation of water vapor and ozone channels, the Airmass RGB's usage at high satellite viewing angels is limited.
The Dust RGB
Information received from the three channels that contribute to the RGB combination
IR 12 - 10.8 (difference of the two IR window channels. Split window) | Information about dust, which gives a high contribution of red component |
10.8 - 8.7 (difference of two IR window channels) | Differentiation between water and ice clouds (channel 07 identifies ice clouds better) |
10.8 (IR Window channel) | Information about cloud top temperatures |
Resulting colors for typical features
Pink | Dust |
Red | Thick ice cloud |
Black | Cirrus |
Orange/yellowish/green | Low cloud |
Typical application areas
- Detection of synoptic, mesoscale and local cloud systems
- Discrimination of different cloud systems and cloud thicknesses (thick - high thin - low cloud)
- Detection of sand storms
- Detection of SO2 plumes from volcanoes
The Day Microphysics RGB
Information received from the three channels that contribute to the RGB combination
VIS 0.8 (visible channel 02) | Surface reflectance (sand, snow), Cloud albedo |
MIR 3.9 (mixed channel, but only solar part is included) | Detection of ice clouds by the differences in particle size and reflectance properties between water and ice clouds, Fire detection |
IR 10.8 (IR window channel) | Temperature of surface/cloud top temperatures |
Resulting colors for typical features
Red | Thick precipitating cloud |
Black brown | Cirrus with thick crystals |
Green | Small ice crystals (often lee cloud) |
Pink | Snow and ice |
Typical application areas
- Convection
- Fog and low cloud
- Fire (contribution from channel 04)
The 24-Hour Microphysics RGB
Information received from the three channels that contribute to the RGB combination
IR 12 - 10.8 | Information about dust, which gives a high value of the red component |
IR 10.8 - 3.9/8.7 | Differentiation between water and ice cloud; better ability to tell the difference between low clouds and fog than the Dust RGB
IR 10.8 - 3.9 is only useful during the night; replacing channel 3.9 with 8.7 is necessary for the 24-Hour Microphysics RGB. It is similar to the Dust RGB, but has a different temperature range. |
From IR 10.8 | Information about cloud top temperatures |
Resulting colors for typical features
Red | Thick ice cloud |
Black | Cirrus |
Ocher/yellowish/green | Low cloud |
Typical application areas
- Synoptic scale and mesoscale cloud systems: convection
- Similar to Dust RGB
The Severe Storm RGB
Information received from the three channels that contribute to the RGB combination
WV6.2 - WV 7.3 (difference of the two WV channels) | Information about moisture content in middle/upper layers of the troposphere |
IR3.9 - IR10.8 | High values of the green component for cold cloud tops, high reflectivity (small ice particles) for very cold cloud tops, and lower reflectivity (normal updraft) |
NIR1.6 - VIS.0.6 | Information about ice and/or water clouds |
Resulting colors for typical features
Blue | Ocean/land |
Red | Convective cloud: large ice particles |
Yellow | Convective cloud: small ice particles |
Typical application areas
- Convective cells
- Detection of severe storm centers