Keith Wood wrote:

Matt Giwer wrote:

I am not an astronomer much less into telescope engineering but I am
certain there are several on earth now that are better than Hubble.
How many does it take to compensate for "wide field, every time" in
terms of cost.

There are no ground-based telescopes that can adequately observe

    Keep in mind I am discussing tradeoffs not trying to make NASA's case.

a wide enough field with AO to be useful for survey work.  The best AO
system can only manage a few arcsecs, as opposed to HST which can
achieve 202 sq arcsecs in one shoot.  Mosaic many ACS images and you
can do very efficient large surveys (e.g. COSMOS) with 50 milliarcsec
resolution.  You could probably do the same on the ground in 1 year if
you had 300 Kecks, or 300 years' of continous observing with 1 Keck...

    I had to think about this one for a while. So a little multiplication says 3600 sq arcsecs per arcmin and 3600x3600 per degree and 3600x3600x180x360 divided by 200 sq arcsecs to survey the sky. It looks like 300 Hubbles also. Obviously it is much faster but prioritization is still going to make tradeoffs. And I would expect the prioritization based upon features much smaller than what is in a wide field.

The lastest 'binocular' one to go on line was something
like $120M and maintenance does not require a shuttle launch but a
van. How much did Hubble cost even in then year dollars? How much will
it cost to get a few more years out of it? What does one shuttle
launch cost today? If on the order of $120 million how about another
binocular telescope instead?

A typical Servicing Mission is on average $500 million.  So yes, you could
buy 4 LBTs or Kecks for that money.  However, that money comes from
NASA's Office of Human Spaceflight, NOT Space Science.  So if the mission
did not take place, the money would be spent on the ISS instead - it would
not be diverted to any other astronomy project.

    So 4 more LBTs can observe the above much smaller objects four times faster than Hubble.

    As to where the money comes from, that depends upon how Congress allocates the money. It is not automatically out of another line item.

It would be different if fixing Hubble this one interim time would
keep it available for decades like Palomar but price it in dollars per
year. That is, how many tens of millions of dollars per year will it cost?

HST's operations cost per year is $300 million, not including
Servicing Mission costs.

    Which is a good number. If a successor is put into orbit then those operations will cost about the same in addition to additional antennas and other base equipment. It is no longer possible to transition the HST operations to the new telescope. Both must be maintained.

If you read the The National Academies
independent assessment of HST science, you'll find that the science
per dollar is very favorable, and that further upgrades (e.g. SM4) are
well worth the money.

    Science per dollar is always favorable by definition as knowledge is priceless, unfortunately, really screws up the spreadsheets.

As far as longevity is concerned, NASA will
keep it running as long as the science community continue to benefit
from it.  If they decide it is no longer delivering the goods, then NASA
will have to abandon HST.

    And that can only be when cheaper methods can gather the same information as there is always a benefit from observation in astronomy. Are not the oldest telescopes still scheduled?

There are no signs of dissent yet from the
science community - they enjoy their new AO toys on mountain-tops,
but still rush to submit proposals for HST observing time nonetheless.
And demand for HST is as fierce as ever, with only 1 in 10 proposals
accepted.

    I would expect a rejection rate about the same for any telescope. And were I in the business I would be tailoring the same idea to each telescope and sending it to all of them. I don't see that as a serious criteria.

And no one is going to argue there are things much better done abover
the atmosphere and there is a replacement tailored to exactly that
planned for launch in 2009 or so. But NASA has finite resources,
finite dollars. Fix Hubble, delay the replacement or delay something
else.

If you are referring to JWST as the replacement, I hate to tell you that
it is not.  JWST is optimized for near and mid IR.  In many ways it is
complementary to HST's UV/Optical capabilities - more of a "follow on"
than a replacement.

    OK, perhaps even better. It gives a new venue of information.

    It does go back to the spreadsheet buster of priceless information. What I have been trying to point out is resources are finite. Someone has to come up with the $300M extra per year to operate this new one and Hubble and the initial costs for new communications hardware and floor space for the personnel. And it has to attract the people to operate it from other professions by outbidding them in salaries. And the non-government telescopes are the ones going to lose personnel. I know of no recruiting drive to get people to major in the required disciples in college instead of non-scientific fields.

    I realize the world is not a zero sum game and know one can know which is the best allocation of resources, that there are only best guesses.

The National Academies: "The Hubble Space Telescope is arguably the most important telescope in history."

As was once Palomar as was once ... you get the picture.

Absolutely!  That's progress for you.  Palomar had it's day, and now it
is HST.  One day, HST will be displaced by a better telescope and that
will take the title as being the most important telescope in history.  But I
can tell you that it won't be a ground-based telescope, and it will not be
JWST.  There are many telescopes which outperform HST in certain
niche areas, but there is no telescope in operation, or planned, that can
beat it at everything.  In that sense, HST is the ultimate 'General Purpose'
observatory that still delivers stunning results, despite its ageing
instruments and equipment.

    Reread what you wrote and read also that Hubble cannot be the best at anything -- visible light at least.

    One can look at this several ways from sentimental value to better the devil we know. I am presenting alternate viewpoints.

    I propose simply that at this very early point in exploiting space it is most profitable to maximize the variation of data rather than to continue to exploit (beat to death) the same spectrum of data.