The infrared spectrum sprawls from the edge of the visible, nominally 0.7 µm, to about one millimeter, such a broad range that it demands subdivision. The definitions of mid- and long-wavelength bands may have grown from the atmospheric transmission windows at 3-5 µm, and 8-14 µm. Those bands generally require different detectors, and also were a handy division between the blackbody peaks of "hot" objects and those of "body-temperature" objects. The strongly absorbed wavelengths in between didn't matter much as long as the infrared was mostly used for looking through the air. Longer wavelengths were lumped as the "far-infrared," a vast region extending to about one millimeter that seemed of little use because atmospheric transmission was spotty and instrumentation was poor.
Atmospheric transmission has not changed, but new infrared detectors and sources have opened up previously little-used regions of the infrared, and satellites have opened the whole infrared spectrum to astronomers. New applications have emerged, such as LWIR monitoring of beehives. That has made drawing dividing lines problematic, particularly on the ends of the LWIR. Should the ends be defined by the atmospheric windows or at some other points? One suggestion was to define each band as an octave wide, spanning a factor of two in wavelength or frequency, but that logical idea failed a crucial practical test because it could not fit both the 3-5 µm and 8-14 µm bands in adjacent octaves. So we're stuck with informal definitions that depend on things like atmospheric windows, and detector ranges, and differ between fields like lasers, astronomy, and night vision.
It could be worse. Geologists built their time scale for the Earth's history on the boundaries between solid rocks, then found that their calendar changed every time a better way was found to date the rocks.
No comments:
Post a Comment