OIG: NASA’s Management of the Stratospheric Observatory for Infrared Astronomy Program

The Stratospheric Observatory for Infrared Astronomy (SOFIA) is an airborne observatory housing a 106-inch telescope
mounted onboard a Boeing 747SP that makes observations from between 38,000 and 45,000 feet, putting it above 99 percent of water vapor that interferes with ground-based infrared observations. The aircraft is operated out of
NASA’s Armstrong Flight Research Center in Palmdale, California, and is currently equipped with six interchangeable
instruments that can be upgraded or replaced. SOFIA is used by astronomers to study astronomical objects and
phenomena including star birth and death, formation of new solar systems, identification of complex molecules in
space, nebulas and galaxies, and transient events like eclipses.

The SOFIA Program is a partnership between NASA—within the Science Mission Directorate’s (SMD) Astrophysics
portfolio—and the German Aerospace Center and operates in cooperation with the Universities Space Research
Association (USRA) and German SOFIA Institute. NASA began development of SOFIA in 1991 and the Program
achieved full operational capability in 2014, 13 years behind schedule with development costs over $1 billion (more
than 300 percent over the initial cost estimate). Because SOFIA has experienced ongoing operational and technical
challenges and has not met science output expectations, NASA has questioned whether its $83 million in annual
operating costs could be put to better use. Likewise, the President’s Budget Request has proposed terminating SOFIA
on multiple occasions including most recently for fiscal year 2021. However, in the past Congress has continued to
fund the Program.

In this audit, we evaluated NASA’s management of the SOFIA Program’s cost and technical performance to ensure
maximum scientific return. Specifically, we assessed the Program’s (1) delivery of reliable flight opportunities;

SOFIA also continues to experience operational and technical challenges related to flight operations, observation
completion, data processing, USRA’s award fees, and instrument development. While SOFIA was designed to fly 960 research (or observation) hours annually, it has yet to achieve this number of hours or dispatch the expected
number of scheduled flights, resulting in much less science output than expected. While SOFIA typically has flown a
10-hour flight profile four times per week, an alternate schedule of 8-hour flights five times per week out of Palmdale
from May to November appears to achieve more time in the stratosphere and therefore higher quality data. Further,
although the Program’s scientific productivity has improved in recent years, the resulting publications and citations are
significantly below what is expected for a mission of SOFIA’s budget and lower than comparable observatories. SOFIA’s
limited science output can be attributed to multiple factors including incomplete observation programs and issues with
the timeliness and quality of data produced during an observation. In addition, performance ratings and award fees for
USRA—which was awarded a contract in 2017 to manage SOFIA Program science operations—have been based on
subjective judgments and were not adequately tied to primary science productivity metrics (publication and citation
rates) that are standard across the astronomical community. The Program has also not developed any new instruments,
only implemented one instrument upgrade, and SMD recently canceled the only new instrument in development due to
cost concerns and technical challenges. Finally, SOFIA has not retired non-productive instruments in a timely manner.

The lack of clear and achievable performance expectations and lack of concurrence between SMD and SOFIA
management on science output goals including publication and citation metrics has reduced productivity and threatens
the Program’s future viability. The Program is unlikely to achieve the community’s expectation of 150 publications per
year by 2022, or the Program’s goal of 100 annual publications, as it only produced 33 publications in 2019 and the
actions proposed to meet this goal fall short of the transformational changes required to address current operational
and technical challenges. Further, the proposed actions are unlikely to mitigate SOFIA’s lack of competitiveness because
of the Program’s poor efficiency on a science-per-dollar basis when compared to other observatories.

To improve the Program’s productivity, we recommended the Associate Administrator for SMD direct SOFIA Program
management to: (1) implement quantifiable research hour requirements; (2) establish a requirement to maximize
research hours in the stratosphere that considers implementing an 8-hour, five times per week flight profile whenever
SOFIA operates from Palmdale between May and November; (3) establish observation program completion metrics,
based on science value prioritization, to increase the probability observations result in publications; (4) develop and
implement a process to increase the timeliness of delivering data to researchers; (5) establish science metrics,

such as publications and citations per year, as criteria for the performance evaluation of the USRA contract award fee;
(6) establish a requirement for new technology implementation; and (7) develop a process for tracking science
instrument productivity and cost to maintain and establish an evaluation program to determine when an instrument
should be retired. We also recommended the Associate Administrator for SMD and SOFIA Program management
coordinate to: (8) reassess SOFIA’s strategy and mission to identify and consider implementing alternative operational
approaches and models to maximize SOFIA’s capabilities within the Astrophysics portfolio and return on investment and
(9) develop consensus between SMD and Program management and implement quantifiable operational and science
output requirements.