IR cameras vs. viewports

[John Brownlee <jonnie@chronic.lpl.arizona.edu>]

	One big question with any CCD or IR  detector setup  is to
carefully consider the scattering and absorbtion regime of the material
you're looking through. I would guess that the 940nm water band would
affect most COTS VNIR IR systems, and that it may be considerably more
difficult to pump out enough light to see in the IR. Also, particulate
sizes in most 'cloudy' water are going to affect the IR pretty heavily,
and then there's the nightmare of detector cooling to consider. Some
recent developments in cryo-pumps may obviate the need forliquid cryogens
on board, it will be interesting to see how this next round of Hubble
upgrades goes. One of the upgrades is to the NICMOS IR camera, which just
ran out of cryogen; they plan on replacing the cryo container with an
electric cryo pump. 
	Interestingly enough, with modern long-wavelength detectors coming
of age (like the recently-declassified sub-millimeter camera), it may
indeed be very posible to pick an optical wavelength regime which gives
long-range visibility in otherwise 'cloudy' water. Once upon a time, I
believe a sub was built using fiber optics instead of viewports called the
DOWB (Deep Ocean Work Boat). It's in the Busby book, I'd be interested in
any accounts of how usable this system was. 
	The wavelength longwards of thermal IR showed a lot of promise in
the literature search I did a long time ago. Wonder if anyone has tried
this sort of system, as it would give better resolution than sonar with
far less of the quirks. Wonder if those long-wave detectors are
commercially available yet? I work mostly with CCD's and InSb IR detectors
nowadays, anyone else up to speed on this?
	Just the same, viewports are still important. I plan on having
them, at least! :)

							John

John Brownlee
Lunar and Planetary Lab
University of Arizona
jonnie @ lpl . arizona . edu