Geostationary Operational Environmental Satellites (GOES) and other satellites

GOES satellites orbit the earth 22,000 miles directly above the equator once every 24 hours, keeping pace with the earth's rotation. Consequently, the satellites appear to remain fixed above the same locations on earth, which is why they are called "geostationary". Each satellite records several images every hour.

The GOES-East satellite orbits above Brazil's Amazon Basin, where it can see both the continental U.S. and the eastern tropical Atlantic Ocean, where hurricanes often develop in summer and fall. GOES-West orbits above the equator over the eastern Pacific Ocean, where it can see (among other things) midlatitude cyclones developing over the ocean south of Alaska (east of Japan) during the winter.

Several other countries operate their own geostationary weather satellites. These other geostationary satellites include Europe's METEOSAT, India's IRS, Japan's MTSAT, and China's FY-2. (For sample images from some of these satellites, see for example NOAA's Geostationary Satellite Server.)

Infrared and visible satellite images

Infrared satellite images record longwave infrared radiation that is emitted by cloud tops, land, oceans, or ice and snow. While most things on earth emit longwave infrared radiation, warmer objects emit more than colder ones. Hence, the relative intensity of infrared radiation emitted by cloud tops, land, oceans and snow/ice gives information about their relative temperatures.

GOES satellites transmit recorded radiative intensities to computers at the National Centers for Environmental Prediction (NCEP) in Washington, D.C. These computers are programmed to create images that display the warmest objects in black and the coldest in white (or sometimes in color). Objects at intermediate temperatures appear in shades of gray. The tops of high clouds are usually (though not always) colder than land or oceans, so the lightest (or colored) parts of a LWIR satellite image usually show high, cold clouds, while the darkest parts of an image usually show land or ocean surfaces, which are usually warmer.

Visible satellite images record visible light (from the sun) that is reflected by clouds, land, oceans or snow and ice. Since visible images record reflected light rather then emitted infrared radiation, they tell us how reflective an object is but nothing about its temperature.

**Note: Since IR radiation is invisible, humans can't see differences in IR radiation intensity (IR "brightness") directly. Instead, a computer is programmed to translate different IR radiation intensities into lighter or darker shades on a satellite image. The translation is arbitrary, but here is how meteorologists have chosen to do it on black and white IR satellite images:

Different temperatures (i.e., different intensities of IR emission) are also sometimes translated into different colors, to create a color enhanced IR satellite image. There are any number of ways to do this—the translation is arbitrary.

Thanks to differences in their temperature or albedo, which give rise to differences in the intensity of radiation recorded by the satellites, the following four features of the earth are generally discernible on GOES weather-satellite images:

land surfaces	snow/ice surfaces
ocean surfaces	cloud tops

We can categorize these four features according to their typical relative albedos and temperatures. (In some cases we have to take into account time of year, time of day, altitude, elevation, or latitude. For example, high-altitude land surfaces are generally relatively cold compared to low-altitude land, all else being equal.)

Features Typically with a Relatively High Albedo	Features Typically with a Relatively Low Albedo
Cloud tops Snow/ice surfaces	Land surfaces (though vegetated areas generally have significantly lower albedo than unvegetated areas, such as deserts) Ocean surfaces (which have the lowest albedo of all except when the sun angle is very low)

Features Typically with a Relatively High Temperature		Features Typically with a Relatively Low Temperature
		Snow/ice surfaces
Land surfaces at low elevations	<--->	Land surface at high elevations
Land and ocean surfaces at low latitudes	<--->	Land and ocean surfaces at high latitudes
Land surfaces in early/mid afternoon	<--->	Land surfaces near sunrise
[Land surfaces in mid/late summer	<--->	Land surfaces in mid/late winter]*
[Ocean surfaces in late summer	<--->	Ocean surfaces in late winter]**

Comparing features on infrared and visible images

On visible images, high and low clouds look similarly bright (because all clouds have relatively similar reflectivities) and are therefore often hard to distinguish from each other. Similarly, clouds and snow-covered ground are often about equally reflective and hence often look the same.

However, fog and low clouds are often easy to see on visible images because they reflect much more light than do nearby ocean or land surfaces.

Of course, visible images can only be recorded during the daytime.

On infrared images, in contrast, fog and low clouds are often hard to distinguish from nearby ocean or land surfaces because these features often all have similar temperatures (though not always--during the daytime, fog reflects solar radiation and so doesn't warm up much, whereas surrounding land surfaces tend to absorb more sunlight and warm up, creating a temperature difference).

However, snow, fog and low clouds are easy to distinguish from high clouds in infrared images because features near the earth's surface and clouds high in the troposphere are usually at different temperatures.

Since clouds, land, oceans and snow and ice never stop emitting longwave infrared radiation, infrared images can be recorded day or night. Land surfaces appear darker (warmer) during the day, lighter (colder) at night.