5

Innovative Cost Reductions for Extended Missions

The committee’s charge included identifying possible innovative ways to reduce costs for extended missions. During the course of this study, the committee heard several presentations addressing cost reduction approaches for extended missions and discussed specific case studies in the search for overarching principles that might be applied to other missions. The committee evaluated approaches to cost savings within the context of increased risk and potential impacts on science return.

COLOCATING OPERATIONS

One method for increasing efficiency for space science missions is colocating multiple mission operations at a given location, which is an approach that NASA already takes for many of its missions. For example, NASA’s Goddard Space Flight Center (GSFC), the California Institute of Technology’s (Caltech’s) Jet Propulsion Laboratory (JPL), and the Johns Hopkins Applied Physics Laboratory (APL) each operate multiple missions using their on-site operations centers. In some cases, these missions are concentrated by type—for example, Earth science missions at GSFC and planetary missions at JPL. However, GSFC also operates the Lunar Reconnaissance Orbiter as well as a number of astrophysics missions, JPL operates some Earth science and astrophysics missions, and APL operates Earth science, heliophysics, and planetary missions. The committee notes that there is no inherent reason that all similar missions have to be handled by the same operations center.

Although colocating multiple missions operations at a single location is likely to produce added efficiencies due to some level of commonality in spacecraft operations, the Science Mission Directorate’s (SMD’s) current portfolio includes competed science missions and principal investigator (PI) teams that provide NASA with different opportunities to draw on scientific expertise that is spread throughout the United States. Added operations efficiencies and scientific synergies may result from colocating science operations and mission operations close to, or at, the host institution for the science team, as exemplified by the Chandra X-ray Center located in Cambridge, Massachusetts, and the Infrared Processing and Analysis Center at Caltech.

INNOVATIVE APPROACHES

The committee also was briefed on the innovative approaches adopted to continue operations during the extended phase of several missions, including the Galaxy Evolution Explorer (GALEX), the Solar, Anomalous, and Magnetospheric Particle Explorer (SAMPEX), and the Mars Exploration Rover Opportunity. The level of NASA support varied for the later stages of these missions, as discussed below, and this factor should be kept in mind when assessing the effectiveness of the approaches.

The GALEX mission provided important ultraviolet astronomy observational capabilities (see Figure 5.1). It transitioned from prime to extended phase in 2007 and was highly recommended in the 2004, 2006, and 2008 Astrophysics Senior Reviews. However, the 2010 Senior Review recommended only 2 more years of operations, followed by close-out. That review also opposed a suggested move of the operations to Caltech, saying that the move would introduce unnecessary risk and would provide no cost savings, given the limited remaining time they were recommending for operating the mission. Subsequently, NASA decided to terminate the mission after just 1 year. The mission PI and the science team negotiated with NASA to transfer operations and ownership of the satellite to Caltech, but several issues arose, including the question of liability associated with possible collisions on-orbit and eventual Earth re-entry. Ultimately, this issue was surmounted by a NASA decision to “loan” the telescope to Caltech, with NASA retaining ownership. However, no NASA funding was provided, so the GALEX team and Caltech endeavored to raise just over $1 million for a bare-bones operation of the satellite for approximately 1 year. Several universities and telescope consortia purchased observing time, JPL funded efforts to complete the galactic plane portion of an all-sky survey, and the PI team raised modest amounts of additional private funding.

Employment of student operators on a part-time basis also reduced costs somewhat. Although these efforts successfully extended the mission, there was no immediate funding or time for science research. According to the PI, the team was exhausted after 1 year, and the satellite was “returned” to NASA and decommissioned. The PI informed the committee that he would not recommend this option to future missions. An unanswered question is the extent to which this approach might have been less taxing on the team, with the possibility of operating in this mode for longer than 1 year, had NASA at least provided partial funding support.

Continuing GALEX operations after the end of NASA funding involved a rather rushed effort with some complicated issues. It is possible that, with more advance notice and careful planning, taking advantage of lessons learned, this kind of effort could be less stressful and more successful in some future situations.

There could be an important ancillary benefit to efforts to transition older missions to a NASA-university/ consortia partnership: increasingly, the development of space hardware and missions is concentrated at NASA centers. Encouraging universities to become involved in extended-phase missions may be one way of rekindling a broader involvement in space hardware and space science. However, this may only be applicable to smaller missions with more focused scientific objectives. Observatories as large and complex as the Hubble Space Telescope and the Chandra X-Ray Observatory cannot easily be transitioned in this way; given the breadth of science that they continue to enable even in their extended phase, it is important that operations do not change drastically.

SAMPEX was NASA’s first Small Explorer mission. Launched in 1992, SAMPEX was designed as a 1-year mission, with a goal of 3 years, to study space weather through measurements of particles and cosmic rays in near-Earth space as a function of solar activity. The mission was extended to cover a full solar cycle, and NASA support ended in 2004. However, data continued to be acquired for another 8 years, with the Aerospace Corporation funding the downloading and Bowie State University operating the spacecraft (starting in 1997) as an educational tool for its students. A GSFC scientist obtained a NASA grant to process the 2004-2012 data and to provide access to the data for the science community. SAMPEX continued to provide valuable science data until it re-entered Earth’s atmosphere in late 2012, just over 20 years after it was launched. Without question, SAMPEX exceeded expectations, thanks in large part to the confluence of factors listed above that enabled the last 8 years of the mission. However, it does not seem realistic to plan future extended missions based on highly uncertain support relying on corporate funding commitments, university interest for educational purposes, or grants that must be competitively secured.

The Mars Exploration Rover has operated on the surface of Mars for more than 12 years (see Figure 5.2). Given that Opportunity’s prime phase was 90 martian days, the duration of the extended phase has exceeded the prime by almost 50-fold. The project has been under continuous pressure during this time to reduce the cost of extended operations without adding risk of loss of mission. The project responded to this new reality by adopting a number of innovative cost-saving measures, most of which were not foreseen at the start of operations, partly due to the very short anticipated prime mission duration. These innovations drew heavily from the actual experience of having operated the spacecraft through the prime mission period. Notable among these cost-saving measures were the use of cloud computing in lieu of purchasing and maintaining hardware systems, the use of information technology automation to handle many routine operational tasks, the cross-training of team members to allow individuals to cover more than one job as extended mission work lessened, and the elimination of deputy positions as team members gained job skills experience and became cross-trained. Overall, this approach was very successful, with increases in efficiency and associated cost reductions implemented “on the fly,” according to one of the mission’s managers.

Unlike the two cases discussed above, NASA did provide continuous, albeit reduced, funding for Opportunity’s extended mission. As noted in Chapter 2, the President’s FY 2015 and FY2016 budget request zeroed out the funding for Opportunity as well as the Lunar Reconnaissance Orbiter, even though both were highly rated in the Planetary Science Division’s 2014 Senior Review. (See Appendix B for sampling of scientific contributions during the extended phases of both missions.) Congress subsequently decided to continue the funding for both Opportunity and LRO.

Subject to recommendations from the Senior Review process, NASA SMD generally expects to extend the mission operations beyond the original prime mission period, provided the spacecraft is returning valuable science data and the cost for extending the operations fits within the program budget. Given that extended operations are

a likely eventuality, planning of the ground system and operations approaches from the early phases of the program can include an awareness (without driving costs) of the potential for a mission extension that is likely to be implemented with reduced budgets, reduced and changing staff, aging hardware, and, possibly, new objectives. Such early steps may provide benefits for later reducing the cost of extended mission operations and limiting the increase in risk.

The committee was briefed on a number of different approaches but did not identify any new over-arching cost-saving principles to apply across the board—every mission has unique circumstances. Using the information presented, the committee was able to extract a number of best practices including the following:

• Allow for the possibility of extended operations without driving costs as projects plan and develop their ground operations and flight procedures for the prime mission.
• Consider the implications of possibly transitioning from prime mission operations into extended missions when recruiting and assigning the operations team for the prime mission.
• Plan for and then cross-train mission and science operations staff to more effectively enable reductions in workforce and staff at reduced risk as a mission transitions to extended phase.
• Perform appropriate trade studies for purchase versus “rental” of computer hardware and data storage (e.g., use of cloud capabilities) for operations and data processing, while addressing factors such as information technology security and upgrade requirements.

The committee determined that communication about Senior Review processes among SMD divisions is relatively good and encourages the divisions to continue this communication about other aspects of extended-mission operations. There are many possible ways that NASA could ensure open communications and dissemination of information, including websites, conferences, and even contractual communications. As the committee has noted, the best time to begin preparations for extended missions is when a mission is still in its formulation phase, a time when decisions can have significant impacts many years after the prime mission has ended.

REPURPOSING EXTENDED MISSIONS TO CREATE NEW SCIENCE MISSIONS

Upon completion of a prime mission and during the transition to an extended phase, opportunities may arise to consider a major redirection of the project. One example is the Deep Impact mission that was launched in 2005 to study the interior of comet Tempel 1. On July 4, 2005, the spacecraft’s impactor collided with the comet, producing effects that were observed by the main spacecraft. Shortly afterwards, Deep Impact’s prime mission ended, even though the spacecraft was still healthy. NASA then sought proposals for an extended mission and eventually selected and merged two proposals that included both original and new members of the Deep Impact team. The extended mission was named EPOXI (Extrasolar Planet Observation and Deep Impact Extended Investigation).

The EPOXI mission recycled the Deep Impact spacecraft to visit a second comet, Hartley 2. The November 4, 2010, flyby of Hartley 2 marked only the fifth time a comet had been visited by a spacecraft. The EPOXI mission flyby revealed that the rocky ends of comet Hartley 2 spew out tons of golf-ball to basketball-size fluffy ice particles, whereas the smooth middle area is more like what was observed on comet Tempel 1, with water evaporating below the surface and percolating out through the dust. Repurposing the Deep Impact spacecraft enabled NASA to take advantage of new ideas and a wider array of expertise that would have otherwise required NASA to initiate and fund the development of a whole new mission.

Another example is the WISE (Wide-field Infrared Survey Explorer) mission, launched in December 2009. WISE surveyed the full sky in four infrared wavelength bands until the hydrogen cooling the telescope was depleted in September 2010. The survey continued as NEOWISE (Near-Earth Object WISE) for an additional 4 months using the two shortest wavelength detectors to detect previously known and new minor planets and to study asteroids throughout the solar system. NEOWISE enabled the discovery of the first known Earth Trojan asteroid. The spacecraft was placed into hibernation in February 2011, after completing its search of the inner solar system.

In response to increasing scientific interest and growing geopolitical concern about the possibility of near-Earth objects (NEOs) impacting Earth and the consequential impacts to human life and damage to the environment and economy, NASA’s Planetary Science Division reactivated the mission (as a directed mission of national priority and no longer subject to the Senior Review process) in December 2013, with the primary goal of learning more about the population of NEOs and comets that could pose an impact hazard to Earth. During its first 3 years of operations, NEOWISE characterized many NEOs and obtained accurate measurements of their diameters and albedos (how much light an object reflects). NEOWISE is equally sensitive to both light-colored asteroids and the optically dark objects that are difficult for ground-based observers to discover and characterize.

As of mid-April 2016, NEOWISE was approximately 73 percent of the way through its fifth coverage of the entire sky. The repurposing of this mission after its prime phase has provided a very cost-effective means of addressing questions of great scientific interest and in this case of great importance to our planet’s, and our own, well-being.

A third example is provided by the Heliophysics THEMIS (Time History of Events and Macroscale Interactions during Substorms) mission. In this instance, a multi-spacecraft mission was partially repurposed to obtain new science. Originally composed of five spacecraft to study magnetospheric substorms, the THEMIS mission proposed that two spacecraft be diverted to lunar orbit. The new mission, called ARTEMIS, has provided important observations of the lunar wake (Wiehle et al., 2011), while the remaining three spacecraft constitute a revised

THEMIS extended mission that continues to provide crucial observations of energy conversion processes in Earth’s magnetotail (Angelopoulos et al., 2013).

RISK ASSESSMENT AND ACCEPTANCE

NASA mission and science operations budgets typically decrease significantly when a mission enters extended phase, which is normally expected and usually justifiable. After that, costs may reduce further as a consequence of additional performance improvements over time and learning-curve effects. However, after several years of extended operations, most missions have implemented all steps that safely can be taken to reduce cost. Further funding cuts increase risk, including a real loss of unique science or possible degradation or loss of a spacecraft. Based on the mission team presentations to the committee, there is a perception among proposal teams that NASA at times may not fully recognize the changed risk posture when reducing funding for mission extensions, instead assuming that funds for extended missions can be continually cut without ramifications. To be fair, NASA is at times under intense budget pressures, and agency officials may believe they have no choice other than to apply such cuts. Moreover, given the national interest needs met by Earth science missions, there is much less risk acceptance for extended missions by the Earth Science Division than the other divisions. Increased risk can take various forms. One example is that missions in extended phase may go for longer periods between communications sessions with ground control. This could mean that a problem on the spacecraft could go undetected and pose a threat to loss of an instrument or the spacecraft. Decisions by NASA and mission proposers to accept such risks have long been made for extended missions, but not everyone involved may be aware of the risks.

THE NEED FOR SUPPORT IN RESPONSE TO SPACECRAFT ANOMALIES

In some instances, mission operations costs may also rise over time due to changes in mission profile; the need to respond to anomalies that are commonly but not always age-related, such as deteriorating performance of flight systems; as well as inflation. For example, the complete loss of one radio receiver on Voyager 1 and the loss of frequency tracking capability on the remaining redundant unit required intense and costly operational workarounds, as did the failure of the high-gain antenna on Galileo during its prime mission phase. Historically, barring such extenuating operational cost drivers, extended missions often experience additional cuts to their budgets at subsequent Senior Reviews, which along with inflation, often result in disproportionate cuts to project-funded science activities. This is because mission management normally prefers to limit increased risk and, therefore, attempts to minimize cuts to the operations budgets. In turn, mission science teams then seek support from research and

analysis programs. However, those programs are also under increasing funding pressure, which means that all-too-frequently science is diminished or sometimes not performed at all.

CONTROL OF COSTS AND RISKS RELATED TO THE INTRODUCTION OF NEW PROCEDURES

In concert with the assessment of past experiences and evaluation of innovative ideas for reducing costs and increasing the science cost-effectiveness of extended missions, the committee discussed the question of increased risk associated with such approaches. It usually costs money upfront to develop new procedures that could eventually reduce costs, but the upfront funding usually is not available during the extended phase of a mission, unless it is diverted from science or essential operations activity. Keeping procedures as simple as possible in the prime mission, which projects should do to the extent possible, may be the best way to control costs and limit risks in extended missions. Increased risk from any new procedure is unavoidable but may be acceptable in some cases. For example, if the alternative is to terminate a mission, then substantially increased risk may be acceptable. Also, risk to the science data is less critical than risk of catastrophic failure of the mission. As is commonly done by project management, all such risks are best identified, described, and carefully evaluated in order to avoid making decisions that could keep a spacecraft operating but drain it of scientific productivity.

DETERMINING THE LIFETIME COST OF SCIENCE MISSIONS

NASA’s present approach is to develop prime mission hardware specifications (e.g., lifetime) such that there will be a high level of confidence in the mission’s ability to meet prime mission requirements. This approach is both understandable and appropriate and has served the agency well. Furthermore, it implies that there is a distinct probability that most missions will survive in good enough shape to propose an extension. Even so, NASA defers formal requests for extended mission operations funding until the approach of the prime mission completion along with achievement of the stated science objectives. This practice probably traces back to the early days of spacecraft development when there was lower confidence that spacecraft and science instrument operations would even reach, let alone exceed, desired mission lifetimes. Some critics have noted that this approach produces life-cycle cost estimates for missions that are lower than they would be if budgets for extended mission operations were included from the start. Moreover, deferring formal requests for mission extensions may encourage some skeptics to question the merits of such extensions. On the other hand, NASA’s 5-year budget projections for the SMD do carry funding for extending missions on a division-by-division basis (sometimes by individual missions and sometimes as an aggregate number), so the planned expenditures are included in NASA’s budget projections.

The committee debated this question and concluded that the current NASA approach is very reasonable. Spacecraft and science instruments are designed and tested for specific lifetimes with corresponding requirements (and associated costs) for component and subsystem reliability. The lifetime design requirements also include margins, which increase the probability that the mission will meet its design lifetime, but do not guarantee how much longer it will continue working beyond its prime phase. After the design lifetime is reached, nobody expects the spacecraft or instruments to immediately stop working, just as nobody expects a household appliance to break the day after its warranty expires, but there is an understanding that degradation in function may occur. The committee also discussed the merits for NASA to further describe this philosophy in its own policy documents as a means to better communicate both internally and externally its intent to extend the operations of missions as long as they continue to return useful data and the resources needed to do so fit within their overarching budget constraints.

In addition, the prime phase of a mission is not only defined by the hardware lifetimes but by the science goals that are to be achieved during that time. If NASA were to define a longer lifetime for a mission from the outset, development, integration, and testing costs would increase, while NASA and the science team might also have to expand the science goals corresponding to a longer prime mission. One of the benefits of an approach that keeps the prime phase separate from the extended phase is that it enables NASA and the science teams to apply knowledge gained during the prime mission to develop expanded, or even totally new, goals for the extended mission. This insight and the new goals cannot be predicted far in advance, so the current approach is a good method of tapping into new knowledge and applying it to an already flying mission.

CONCLUSION

The committee is very supportive of the current NASA approach to mission design, which provides a high probability of achieving prime mission objectives while also allowing a reasonable likelihood that an extended phase with high science return will be achievable. As stated earlier, extended missions enable new science, provide for data continuity, and enable long baseline studies—all at very modest incremental cost. The committee has identified a number of good/best practices for missions to adopt in order to limit increased risk and prepare to operate extended missions under likely reduced budgets. Various cost-saving approaches were presented to the committee, and a number of positive attributes were identified, although no global solutions were found, given the distinct aspects of the various missions. The committee is supportive of the acceptance of increased risk during the extended phase of most missions while noting that the national interests or needs aspects of Earth science missions (and possibly some Heliophysics missions as well) establish different risk acceptance levels. The committee also notes the importance of considering operations trades along with science impacts when budget reductions are required and notes the importance of providing roughly constant funding for highly performing missions after the first two Senior Reviews.