
The spectral range of the Landsat Thematic Mapper (TM) instrument

TM images of San Franciso Bay in all 7 TM spectral bands

Combining TM bands 5, 4, & 2 to make an image

TM 542 image of San Francisco Bay (combined with terrain data)

TM 543 image of San Francisco Bay

TM 432 ("False color infrared") image of San Francisco Bay

TM 321 ("Natural color") image of San Francisco Bay

TM 753 image of San Francisco Bay
These images are compressed versions of high definition television (HDTV)
images showing how Landsat data, which spans a very broad swatch of the
electromagnetic spectrum, can be turned into images. The TIFF versions of
these images are full resolution HDTV frames (1920 x 1080). All images
have the HDTV standard aspect ratio (16:9).
The Thematic Mapper (TM) on Landsat 4 and 5 observes reflected sunlight
from the Earth all the way from blue in the visible part of the
electromagnetic spectrum to shortwave infrared well beyond the ability of the
human eye to percieve. The TM instrument also can observe infrared radiation
actively emitted by the Earth from thermal infrared radiation. Landsat 7
carries an improved version of the TM instrument, called ETM+. In addition to
7 channels of spectral data collected by the older TM instruments, ETM+ can
observe in a special panchromatic band spanning the entire visible spectrum at
twice the resolution of the TM bands (15 meter resolution instead of 30
meters). The ETM+ also has a major improvement in the resolution of the
thermal band (60 meter resolution instead of 160 meters).
A standard way to create images from raw Landsat TM and ETM+ data is to
display a single band as a primary color, then combine different bands to
create a full color image. Images shown here demonstrate combining three
bands to make a color image using TM bands 5, 4, & 2, which covers a
very broad range of the TM's spectral coverage. It is also shown in
combination with a digital elevation model. Terrain data is shown with
vertical features exagerated by a factor of three to emphasize details.