Order Code RS21767
Updated June 29, 2006
CRS Report for Congress
Received through the CRS Web
Hubble Space Telescope: Should NASA
Proceed with a Servicing Mission?
Daniel Morgan
Analyst in Science and Technology Policy
Resources, Science, and Industry Division
Summary
The National Aeronautics and Space Administration (NASA) estimates that
without a servicing mission to replace key components, the Hubble Space Telescope
will cease scientific operations in 2008 instead of 2010. In January 2004, then-NASA
Administrator Sean O’Keefe announced that the space shuttle would no longer be used
to service Hubble. He indicated that this decision was based primarily on safety
concerns in the wake of the space shuttle Columbia accident in 2003. Many critics,
however, saw it as the result of the new Vision for Space Exploration, announced by
President Bush in January 2004, which focuses NASA’s priorities on human and robotic
exploration of the solar system. Hubble supporters hope to reverse the decision and
proceed with a shuttle servicing mission. Dr. Michael Griffin, who became NASA
Administrator in April 2005, has stated that he will reassess whether to use the shuttle
to service Hubble after there have been two successful post-Columbia shuttle flights.
Problems during the first flight in July 2005 led to postponement of the second, which
is now planned for July 2006. This report will be updated.
Introduction
NASA launched the Hubble Space Telescope in 1990 aboard the space shuttle
Discovery. Unlike other NASA space telescopes, Hubble was designed to be serviced
regularly by astronauts. That design proved fortuitous when it was discovered that
Hubble had a defective mirror that produced blurry images. Astronauts on the first
servicing mission in 1993 were able to install corrective optics, allowing years of
scientific accomplishments and generating widespread scientific and public support.
Additional servicing missions were conducted in 1997, 1999, and 2002 to replace aging
hardware and install advanced scientific instruments. Two more shuttle missions to
Hubble were scheduled: another servicing mission in 2004 (known as SM-4) and a
retrieval mission to bring the telescope back to Earth in 2010. Following the space shuttle
Columbia accident in February 2003, however, then-NASA Administrator Sean O’Keefe
decided not to proceed with either flight. Current NASA Administrator Michael Griffin
plans to revisit that decision once the shuttle returns to regular flight.
Congressional Research Service ˜ The Library of Congress

CRS-2
The Hubble Space Telescope
Roughly the size of a school bus, the Hubble Space Telescope was designed to make
astronomical observations of the universe in the visible, ultraviolet, and near-infrared
wavelength bands. Although ground-based telescopes can also make visible and infrared
observations, Hubble’s location above the Earth’s atmosphere enhances image clarity,
enabling astronomers to look at fainter, more distant objects and further back in time.
Hubble is operated for NASA by the Space Telescope Science Institute (STStcI) near
Baltimore, MD. Websites maintained by STScI ([http://hubblesite.org/]) and NASA
([http://hubble.nasa.gov]) provide information about the telescope and its discoveries.
During servicing visits to Hubble, shuttle crews repair or replace aging equipment
and install updated scientific instruments. Hubble has six gyroscopes for pointing the
telescope, but two are now nonfunctional and one has degraded performance. Until
recently, three were required to achieve the accuracy needed for scientific observations.
New techniques implemented in August 2005 allow operation on just two gyroscopes, so
that one is kept in reserve. Solar arrays generate electricity for the telescope; the energy
is stored in batteries. Hubble has no propulsion system, relying instead on the space
shuttle to boost its orbit so that it does not reenter Earth’s atmosphere. The tasks for the
SM-4 mission included replacing all the gyroscopes and batteries and a fine guidance
sensor, emplacing new thermal protection blankets, boosting Hubble’s orbit, and
installing two new scientific instruments (the Cosmic Origins Spectrograph for ultraviolet
observations of chemical composition, and the Wide Field Camera 3 for observations
from ultraviolet through near-infrared).
Hubble was designed to operate for 15 years, a milestone that was reached on April
25, 2005. NASA had planned to extend the operational period until 2010, at which time
the space shuttle would bring Hubble back to Earth to prevent an uncontrolled reentry that
could pose a debris risk to populated areas. Funding constraints, however, led some
NASA officials to conclude that Hubble’s operations should be brought to end earlier than
2010. The “funding wedge” created by ending the servicing missions would be used to
build the new James Webb Space Telescope (JWST), which is being designed for infrared
observations. The debate over how long to continue to operate Hubble, and linkage with
funding for the JWST, was under way at the time of the Columbia accident.
The Columbia Accident and the Decision to Cancel SM-4
On February 1, 2003, the space shuttle Columbia disintegrated as it returned to Earth
following a 16-day scientific mission. All seven astronauts aboard perished. The shuttle
system was immediately grounded. NASA established the Columbia Accident
Investigation Board (CAIB) to determine the causes of the accident and recommend
corrective actions. (CRS Report RS21408, NASA’s Space Shuttle Program: The
Columbia Tragedy, the Discovery Mission, and the Future of the Shuttle
, by Marcia S.
Smith, discusses the Columbia accident.)
It was quickly apparent that SM-4 would be delayed. As CAIB deliberated, NASA
decided that the 2010 Hubble retrieval mission was too risky compared to the benefits.

CRS-3
The agency began investigating alternatives for a controlled deorbit of the telescope so
that any debris would fall in an unpopulated area such as the Pacific Ocean.
On January 16, 2004, Mr. O’Keefe informed workers at STScI and NASA’s Goddard
Space Flight Center (which built Hubble and oversees STScI) that he was canceling
SM-4. According to the director of STScI, Dr. Steven Beckwith, Mr. O’Keefe cited
several factors: the shuttle would not have the ISS as a safe haven; the changes required
to meet the CAIB’s recommendations regarding non-ISS related shuttle missions would
not have application beyond the servicing mission, making their expense questionable;
completing ISS construction by 2010 will require all the shuttle flights in that time period;
Hubble’s life would be extended for only a few years; and astronomers have other ground-
and space-based telescopes they could use.1
Reaction to the Decision
Two days before Mr. O’Keefe announced his decision, President Bush directed
NASA to embark on a new exploration initiative, requiring a shift in program and funding
priorities. (See CRS Report RS21720, Space Exploration: Issues Concerning the
“Vision for Space Exploration”
, by Marcia S. Smith.) Funding for the new initiative
would come primarily from canceling, deferring, or delaying other NASA programs.
Although Mr. O’Keefe stated that the Hubble decision was based primarily on shuttle
safety concerns, the timing of his announcement led many commentators to conclude that
it was linked to the priority shifts required by the President’s initiative. While some
media accounts praised the NASA Administrator for making a difficult decision, others
called Hubble the first “victim” of the President’s initiative and chided NASA for putting
the new exploration goals ahead of the astronomical research performed with Hubble.2
Servicing Options
Initial opposition to the cancellation of SM-4 focused on attempts to reverse Mr.
O’Keefe’s decision and proceed with a shuttle mission. The debate centered on
comparing the risk of a mission to Hubble with the risk of a mission to the ISS. Shortly
after the cancellation decision, NASA’s then-Chief Scientist, Dr. John Grunsfeld, an
astronaut who was a member of the 1999 and 2002 Hubble servicing crews, commented
that if a shuttle mission to Hubble were mounted, it would be necessary to have a second
shuttle ready to launch in case the first one encountered difficulties.3 NASA had a backup
shuttle ready for the first shuttle launch after return to flight, and plans to have one for the
second, but not for subsequent missions if safety modifications appear to be working well.
Attention soon shifted to robotic servicing options, which dominated the public
discussion of Hubble’s future throughout most of 2004. At a Senate Appropriations VA-
1 See [http://www.stsci.edu/resources/sm4cancellation/sm4meeting.html].
2 For example, see “Sacrificing Hubble,” Space News, Jan. 26, 2004, p. 12; and Brian Berger,
“Canceled Hubble Repair the First Victim of New NASA Vision,” Space News, Jan. 26, 2004,
p. 6.
3 Quoted in Brian Berger, “Canceled Hubble Repair the First Victim of New NASA Vision,”
Space News, Jan. 26, 2004, p. 6.

CRS-4
HUD-IA Subcommittee hearing on March 11, 2004, Mr. O’Keefe agreed with a request
from Senator Mikulski to ask the National Research Council (NRC) to study options for
extending Hubble’s life, including both shuttle and robotic missions.
In November 2004, Congress passed the FY2005 Consolidated Appropriations Act
(P.L. 108-447), which provided $291 million for a Hubble servicing mission. The
conference report (H.Rept. 108-792) stated that “a successful servicing mission to Hubble
should be one of NASA’s highest priorities.” The report language did not specify whether
the servicing mission should involve the space shuttle or a robotic mission.
The final NRC report on servicing options, released on December 8, 2004, found it
“unlikely that NASA will be able to extend the science life of [Hubble] through robotic
servicing” and that the risk of a shuttle mission to Hubble is similar to the risk of a single
shuttle mission to the ISS. The report recommended a shuttle servicing mission, and a
robotic mission only for deorbiting the telescope at the end of its useful lifetime.4 Robotic
servicing now appears to be off the table.
Shuttle Servicing Revisited
Dr. Michael Griffin was sworn in as NASA Administrator, replacing Mr. O’Keefe,
on April 14, 2005. At his Senate confirmation hearing on April 12, 2005, Dr. Griffin
stated that he was very familiar with the robotic servicing option and did not think it was
feasible within the required time frame. He added that he will revisit the question of
whether to use the shuttle to service Hubble after the second successful post-RTF shuttle
flight, at which time NASA will be able to assess the risk factors associated with
“essentially a new vehicle”. The NASA Authorization Act of 2005 (P.L. 109-155),
enacted in December 2005, calls for a shuttle servicing mission after the shuttle returns
to flight successfully “unless such a mission would compromise astronaut safety” and
requires a status report on servicing plans within 60 days of the landing of the second
successful post-RTF flight. Problems during the first post-RTF flight in July 2005 led to
postponement of the second, which is now planned for July 2006.
Because Dr. Griffin’s decision to reconsider a shuttle mission to Hubble came after
the release of the FY2006 budget, its cost was not included in the budget request for that
year. A budget amendment submitted on July 15, 2005, requested an additional $30
million for Hubble, but this was only to preserve the option of a shuttle servicing mission,
and was not intended to reflect the full cost of servicing Hubble if a decision is made to
proceed. In November 2005, Congress appropriated a total of $271 million for Hubble:
the original request, plus the addition requested in the July budget amendment, plus
another $50 million “to continue planning, preparation, and engineering activities” for
SM-4, “pending a final decision” to proceed (H.Rept. 109-272, conference report for P.L.
109-108). The 2005 authorization act (P.L. 109-155) did not specify Hubble funding.
The FY2007 budget request states that it includes full funding for a shuttle-based
servicing mission to Hubble in early FY2008, but it does not specify a dollar figure. The
total FY2007 request for Hubble, including ongoing operations as well as preparations for
4 National Research Council, Assessment of Options for Extending the Life of the Hubble Space
Telescope
, National Academies Press, 2005. Online at [http://books.nap.edu/catalog/11169.html].

CRS-5
a servicing mission, is $336.7 million. According to the Government Accountability
Office (GAO), NASA estimated in late 2004 that a shuttle servicing mission to Hubble,
including the full cost of a shuttle launch, would cost between $1.7 billion and $2.4
billion. GAO described this estimate as “not yet definitive”, however.5 It is considerably
more than pre-Columbia estimates that were in the vicinity of $100 million plus the
marginal cost of a shuttle launch.
Deorbiting Hubble
Plans for deorbiting Hubble also remain uncertain. Before the Columbia accident,
the space shuttle was to return Hubble to Earth at the end of its lifetime, but there is no
longer any expectation of retrieval by the shuttle, even if there is a shuttle servicing
mission. Hubble has no propulsion system of its own, however, so a propulsion module
must be attached to the spacecraft to permit a controlled deorbit. Remotely attaching the
required new hardware is technologically challenging. If there is a shuttle servicing
mission, a deorbit module could be attached at that time. However, analysis by NASA
during 2005 indicates that Hubble is unlikely to make an uncontrolled reentry until at least
2020, rather than 2012 as previously believed, and the agency now considers the
deorbiting issue to be “beyond the budget horizon.”6 Boosting Hubble’s orbit during a
shuttle servicing mission could delay the date of reentry even further, but at some point,
deorbiting will be necessary if an uncontrolled reentry is to be avoided.
Key Questions for Congress
If there is no successful servicing mission, Hubble’s lifetime is likely to be limited
by the failure of its gyroscopes and batteries.7 Both systems wear out over time, and
historical experience permits projections of their future performance. The NRC study
estimated that science operations would likely cease in about September 2007 as the result
of gyroscope failures. Now that Hubble is operating on two gyroscopes rather than three,
NASA believes that this date can be extended until mid-2008.8 Servicing could still
restart operations after science operations cease. However, the NRC also estimated that
battery degradation would result in permanent structural damage in about May 2011, and
perhaps as soon as December 2009 in the worst case. (A minimum level of battery power
is required to keep Hubble warm and prevent warping of the optical system.) Servicing
after this date would not be successful. NASA estimates that if left unattended, Hubble
will make an uncontrolled reentry in about 2020.
In June 2006, it seemed that Hubble science might be limited sooner than expected
by problems with Hubble’s main camera, the Advanced Camera for Surveys (ACS). The
ACS was taken off line on June 19, 2006, to investigate problems with its power supply
5 Government Accountability Office, Space Shuttle: Costs of Hubble Servicing Mission and
Implementation of Safety Recommendations Not Yet Definitive
, GAO-05-34, Nov. 2004.
6 NASA FY2007 budget justification, p. SAE SMD 3-20.
7 In some scenarios, the limiting factors might instead be the failure of Hubble’s fine guidance
sensors or its avionics system.
8 “Hubble Space Telescope Begins ‘Two-Gyro’ Science Operations,” NASA press release,
August 31, 2005.

CRS-6
electronics. NASA now anticipates, however, that these problems will be fully resolved
in early July 2006.9
As Congress continues to review the Hubble servicing issue, the following questions
may arise:
! How much would a shuttle servicing mission cost? What other programs
would be cut to pay for it? (For example, to offset the funding that
Congress provided for Hubble in FY2005, NASA’s May 2005 operating
plan postponed two other astronomy missions and reduced funding for
Mars exploration.) What is the risk to the shuttle astronauts? Is the
science to be gained worth the cost and risk?
! If Hubble is serviced, it is expected to operate for about five more years,
rather than being deorbited in 2010 as previously planned. How much
will operations after 2010 cost? (The FY2007 budget request, which was
the first to extend its out-year estimates to FY2011, estimated Hubble
costs in that year of $138.5 million.) How might those costs affect
funding for other astronomy missions, such as JWST?
! To what extent could astronomers use ground-based telescopes instead
of Hubble? To what extent could they use other space telescopes, such
as JWST or the Spitzer Space Telescope? To what extent could they rely
on analyzing the archive of data already collected by Hubble?
(Approximately five years of Hubble data have yet to be analyzed.)
9 “NASA Issues Hubble Space Telescope Status Report,” NASA press release, June 27, 2006.