BRUCE.MACINTOSH@BONNIE.ASTRO.UCLA.EDU (04/25/91)
From: BRUCE.MACINTOSH@BONNIE.ASTRO.UCLA.EDU >[Question about the feasibility of using the Hubble space telescope for > reconnaisance] The moderator implied this has been discussed before, but since I didn't see that discussion I feel free to discuss it again. This is very unlikely to occur, and wouldn't be very feasible, for several reasons. (1) The military has several satellites of its own, of similar basic configuration and capability, but with sensors, optics and guidance optimized for the reconnasiance role. (2) The HST instruments don't seem to me to be well-suited to military imaging. The most obvious choice would be the wide-field/planetary camera. The current WF/PC (it will be the first instrument to be replaced) has two imaging modes: in one, the field of view on the ground would be 232 meters on a side, with a pixel size of 0.15 meters. In the other, field of view would be 100m with a pixel size of 0.06 meters. [1] Whether these would be useful would depend on the exact mission, but I certainly can't think of any military missions being well-served by an image covering such a small area. (The other imaging instrument, the faint-object camera, would have a field of view 16 meters on a side with pixels covering 3cm.) (3) The HST control software has a very large number of safeguards to keep the telescope from pointing at the Earth. It would have to be largely rewritten, a time-consuming and difficult task. (4) The reason the software has all those safeguards is that the Earth (by daylight, at least) is far too bright to even consider looking at. The instruments are designed for looking at extremely faint astronomical objects, not daylight scenes on the Earth. I don't know the exact minimum exposure time, but it is presumably set by the mechanical capability of the camera shutter and would be on the order of 0.1 seconds. Looking at daylight Earth, even in high-resolution mode, the image would be completely saturated (overexposed) in this short a time. (Camcorder CCDs and presumably military ones read out much faster than this, but astronomical CCDs read out in a "slow scan" mode to reduce noise; reading out the WF/PC CCDs probably takes about 1 second. The control electronics in the WF/PC are all ROM, so this can't be changed.) The faint-object camera is designed to detect individual photons - it would saturate completely looking at the earth, and probably be damaged. I believe several other instruments and portions of the telescope could be damaged by direct exposure of the daylight earth. I am less sure how useful the HST would be for imaging the night-time earth (by man-made light and starlight.) A 100-watt lightbulb from 300km up is about 6 or 7th magnitude; this is actually still to bright for the HST to image.) On the whole, it wouldn't work out very well. Footnote: [1] Figures are based on a 300-km orbit for the telescope; embarrasingly, I can't remember the exact orbit. Resolutions for the cameras *are* exact, taken from "Electronic and Computer-Aided Astronomy" (Ian McLean 1989; an excellent book for explaining how CCD cameras and modern infrared imaging detectors actually work.) The aberration in the primary mirror isn't a major factor, as the telescope can be focused so images have "bright cores" close to the design performance of the telescope, at the cost of a lot of scattered light; since the daylight Earth is too bright anyway, the lost light isn't a major problem. Bruce Macintosh Imaging Infrared Detector Laboratory, Department of Astronomy, UCLA bruce@bonnie.astro.ucla.edu