Posted by Peter Eisenhardt on June 17, 1999 at 15:03:33:
Given the large archive of data in F110W and F160W, you probably need to repeat these. It would also be very useful to have a filter which matches groundbased J and H - 165M is not bad for the latter, but 110W is not a good match to J. I expect that an IR version of Gunn z would have a lot of utility and much better performance than on a CCD, allowing one to search for z=8 Gunn z dropouts, for example. Also, consider an open (0.9 - 1.9um) IR filter which would complement the STIS open filter (perhaps there should be a 0.2 - 0.9 open filter on the optical side of WF3) to provide very deep images with some color info. Finally, on the IR side it makes sense to have filters tuned to what we can't get from the ground, i.e. atmospheric absorption bands between J and H and perhaps H and K (though the latter would argue for extending the response to 2um, which is problematic with the warm HST optics.) Given that HST isn't that big and that adaptive optics is coming along, in general I would try to emphasize what can't easily be done from the ground, i.e. wavelengths blocked by the atmosphere; and high resolution at the shorter end of the IR range, where adaptive optics is more difficult and diffraction is less of a problem for HST.