As noted in Chapter 1, user demands and the expansion of the Space Surveillance Network are increasing the complexity of the space situational awareness mission. The 2010 National Space Policy provides a de facto context for the role of the Joint Space Operations Center (JSpOC) as an arm of the Department of Defense and the U.S. government, including the responsibility for “[d]evelop[ing], maintain[ing], and us[ing] space situational awareness information from commercial, civil, and national security sources to detect, identify, and attribute actions in space that are contrary to responsible use and the long-term sustainability of the space environment”; for “[m]aintain[ing] and integrat[ing] space surveillance, intelligence, and other information to develop accurate and timely space situational awareness”; and for “[using] space situational awareness information … to support national and homeland security, civil space agencies, particularly human space flight activities, and commercial and foreign space operations.”1 Whereas the JSpOC’s duties are continuous and long term in nature, programmatic funding is available only on an annual basis and is unpredictable. Implementation of the 2010 National Space Policy will likely increase the demands placed on the JSpOC’s programs. Funding levels will likely have to increase as well to support these increasing demands.
This chapter explores some of the broader issues that will be encountered as the Air Force Space Command (AFSPC) moves forward. These include development of strategic analysis and vision, relationships with the external space community and finally, costs and risks.
Air Force Space Command and the JSpOC could benefit from a strategic analysis of space situational awareness-related activities, including algorithm and model development and upgrades as well as operations and personnel management. Lacking such an analysis could lead to making budget decisions and model development priorities in an uncoordinated way, rather than through a coherent, well-thought-out, and prioritized strategy. However, with such an analysis completed, as personnel changes occurred or if funding increased or decreased over time, a priorities-based strategy could be developed to provide guidance as to how efforts could be structured and
1 National Space Policy of the United States of America, June 28, 2010, available at http://www.whitehouse.gov/sites/default/files/national_space_policy_6-28-10.pdf.
resources allocated according to priorities (fiscal as well as personnel). A useful strategic analysis would address the following four major items:
• An assessment of the current state of affairs.
• Specific goals in each of the major space situational awareness areas, along with concrete metrics for success and associated timelines over, for example, 5 to 10 years.
• The funding required to meet the goals laid out above, detailed with clear priorities identified in case full funding cannot be provided by the U.S. Air Force (USAF).
• A means of assessing progress in achieving AFSPC objectives. This will allow AFSPC to gauge its ability not only to meet the original strategic goals but also to respond dynamically to the ever-changing space environment.
A good strategic analysis will result in a vision that will provide not only a roadmap of what an organization intends to accomplish over a reasonable period of time (probably 5 to 10 years), but also information on how these activities will be completed. The vision would include short- and long-term objectives, a schedule of benchmark achievements to be accomplished, and priorities among them. Noting budget realities within the plan will help focus its priorities. Regular meetings involving JSpOC stakeholders and users would be part of the vision’s development, execution, and review, providing opportunities for the JSpOC’s activities and work to be communicated to the space community, while also providing for any algorithm and model development or upgrades to be peer reviewed and validated. The vision would be revised and updated at regular intervals, as would requirements and assessments of how well the JSpOC is meeting them.
Recommendation: The Air Force Space Command should conduct a strategic analysis of its space situational awareness-related activities, particularly as they pertain to the directives of the 2010 National Space Policy.
Recommendation: The Air Force Space Command should further develop the vision for the future of the Joint Space Operations Center Mission System and the system of systems based on the 2010 National Space Policy.
Beyond the improved hardware and software infrastructure support that the JSpOC Mission System (JMS) represents, the committee also found that cultural changes within AFSPC will be necessary for the continued and future success of the JSpOC enterprise. There is currently a significant disconnect between advanced research in the field of space situational awareness, writ large, and the actual practice enshrined in the current AFSPC standardized astrodynamics algorithms. A similar disconnect exists between the activities and needs of the user community and the activities of AFSPC and the JSpOC. This indicates that a culture at AFSPC has developed that is isolated from current research trends and advances, meaning that the knowledge and even vocabulary of key players within AFSPC have in many instances diverged from those of the research community over time. This disconnect not only inhibits AFSPC from taking advantage of new ideas and processes that could improve the system and potentially increase efficiency, but also means that AFSPC becomes unable to describe its current practices in terms of commonly used terminology within the larger space community. This divide cuts both ways and also prevents the user community from properly understanding and interpreting the information that is distributed by AFSPC and the JSpOC. These disconnects can be traced back to a lack of proper documentation of the AFSPC algorithms, the absence of external technical peer review of AFSPC activities, and the lack of direct interaction between the larger user and research communities and AFSPC. This divergence between AFSPC and the user and research community will only grow if it is not appropriately addressed.
A key component of these disconnects is a lack of peer review of technical developments and activities. While some groups and individuals within AFSPC might publish their work through a rigorous peer review process, the
more common approach for the dissemination and critique of algorithmic and/or modeling developments appears to be through networking at conferences and working within closed topical area communities. Thus, it appears that there is little peer review by knowledgeable individuals outside the AFSPC community. Lack of peer review means that:
• The space-faring community is not able to obtain a detailed understanding of AFSPC algorithms that would allow them to better use its products.
• AFSPC does not receive the many benefits that are typically associated with the review of technical work by external subject-matter experts.
Finding: The modelers and algorithm developers who support the JSpOC mission have developed an internal community that lacks sufficient two-way interaction with the larger research and user community.
Recommendation: Joint Space Operations Center algorithm and model developers should fully communicate the results of their work and their development activities, such as in appropriate peer-reviewed publications and conferences, so that users gain greater insight into and understanding of the underlying assumptions associated with catalog activities.
It is important for AFSPC to foster relationships with a variety of external communities to improve transitioning of new technologies from research to operations. These communities include, for example, the astrodynamics community (at large and within, for example, the American Astronautical Society and the American Institute of Aeronautics and Astronautics), the space weather and atmospheric modeling communities, the software and hardware technology communities, and the statistical estimation community. Open communication with these and other communities will keep AFSPC informed about the latest research and development advances that directly relate to space situational awareness, the methods and approaches used by other organizations and developers who are attempting to solve similar challenging problems, and other software and computational tools and capabilities that might be applicable to space situational awareness activities. The committee notes that although the AFSPC charter is fairly unique, appearing to have no parallel in the world, many of the challenges that it faces (sparse data sets, limited modeling capability, strong demands for timeliness, sensitive capabilities) also appear in the financial, petrochemical, and medical industries, in meteorology, and in other areas. A survey of these industries, and interaction with them, may yield ideas (or at least lessons learned) for alternative integrated solutions to the challenge at hand.
The pathway toward remedying this problem must provide AFSPC, JSpOC, user, and research communities appropriate forums in which issues of interest can be discussed and shared. Several different approaches could help break down the current walls to communication and understanding. Approaches can be divided into two main categories: allowing outside comments, ideas, and technology infusion into AFSPC, and communication of relevant internal AFSPC activities to the outside. To successfully bring these communities together, both of these categories of approaches must be pursued.
Some simple guidelines and processes can be outlined for such exchanges of ideas and for technology infusion to occur. To be effective they should involve some formal processes for new technology infusion, avenues for proposing new capabilities, and identification of appropriate methodologies for making updates to the system. Without such an infrastructure in place, JMS will fall far short of its potential and intended capabilities. These activities are compressed into the following two main needs:
• Process to identify necessary improvements and capabilities.
—Creation of an external advisory group, including technically knowledgeable members from the research community, to suggest specific improvements, motivate new developments, and monitor progress as JMS evolves.
—Creation of a process for internal users and analysts to identify, suggest, and develop improvements.
—Means for making research funds competitively available for developing and raising the technology readiness level for possible improvements.
• Process to incorporate improvements and capabilities.
—Development of a specific methodology and formal mechanisms for how possible improvements to algorithms/models/methods/data/capabilities will be investigated and ultimately evaluated, incorporating peer review at some stage.
—Development of a specific set of architecture rules for how new capabilities can interface with the existing tools.
—Identification and communication of rigorous testing protocols for allowing new algorithms and capabilities into the system. Development of a database containing a sufficient amount of sensor tracking data that could be accessed by the research community for the development and validation of new algorithms that support space situational awareness.
Given the status of the JSpOC as the premier organization for space surveillance and space situational awareness and the international cooperation mandated by the 2010 National Space Policy,2 the committee believes that a path of continual improvement, including a process for evaluating, testing, and implementing improvements, is necessary. An advisory group that includes non-USAF members and meets regularly can be a mechanism to review requirements, assess whether updates or upgrades to requirements or models are needed, and assess how well the JSpOC is meeting the goals outlined in its strategic vision.
Recommendation: While recognizing security issues, the Air Force Space Command (AFSPC) should become more open and transparent in the creation and dissemination of its algorithms and products. Specifically:
• The newly created AFSPC Astrodynamics Advisory Committee should be modified to include a balance of internal (e.g., Air Force Research Laboratory, Defense Advanced Research Projects Agency, Missile Defense Agency, etc.) and external subject-matter experts to encourage the introduction of new approaches and new ideas. Examples of external members include representatives from other federal agencies (e.g., the National Aeronautics and Space Administration, the National Oceanic and Atmospheric Administration, the National Reconnaissance Office, etc.), research centers (such as federally funded research and development centers), commercial industry, and academia.
• AFSPC should create a process and an infrastructure to identify and incorporate improvements into the Joint Space Operations Center (JSpOC) and a way to evaluate candidate improvements (e.g., testbeds, benchmarks).
• AFSPC should expand opportunities for astrodynamics and computation specialists to participate in improving the algorithms used in the JSpOC Mission System. This expanded participation should be achieved by advocating for research initiatives and engaging members of the research community to serve as peer reviewers, and by appropriate sharing of data.
• The JSpOC should provide a database containing a reasonable amount of sensor tracking data that would be available to the research community for the development and validation of new algorithms that support space situational awareness.
Bringing the AFSPC culture into closer contact with the larger research and user communities will help the JSpOC to maintain its position as the premier organization for space surveillance and space situational awareness. Of course, peer review of work performed by AFSPC may not be appropriate in all situations; for instance, security concerns may prevent peer review by the uncleared community (i.e., persons not possessing security clearances), and so workshops may sometimes be better suited in such cases for the presentation of developments or advances. However, peer review can still exist in these cases and represents an effective method for ensuring that best practices are pursued, developed, and maintained. Adoption of peer review by AFSPC will help ensure the proper application of scientifically tested and accepted results, underlying intricacies, and other key developments.
2 National Space Policy of the United States of America, June 28, 2010, available at http://www.whitehouse.gov/sites/default/files/national_space_policy_6-28-10.pdf.
As AFSPC moves forward into the JMS era, there will be many benefits, but also risks and potential additional costs. These costs and risks will be both in the JMS community, as it tries to select new algorithms to incorporate, and in the legacy user community, as it tries to keep up with JMS changes.
The current program to develop the JSpOC Mission System needs to ensure that algorithms being considered for inclusion in the new system are fully evaluated so that an informed decision can be made about whether to include them in the operational system. Although the legacy algorithms in the previous Space Defense Operations Center (SPADOC) and the astrodynamics support workstation systems offer a low-risk, proven option for operations, JMS should make sure that potential opportunities for improvements are not overlooked. Also there may be new requirements for which legacy algorithms are not available. The process for evaluating new algorithms should be used on a continuing basis even after JMS is completed, in order to appraise emerging technologies.
There has been a historical tension between the need for an operational system to have a stable, interoperable algorithm base and the wish to frequently upgrade to new technologies that seem promising. Sometimes the early promise of an improvement disappears in the face of operational data and constraints. To minimize risk and cost, it is critical to have a process that will give assurance that a new algorithm will work in an operational environment. Part of this process would be a set of improvement or goal criteria that can be measured and evaluated. Formal requirements documents can be used, but they often do not contain the detailed information that the evaluation process would require (e.g., is a proposed modification acceptable that improves run times by 10 percent but worsens accuracy by 5 percent?). Hippocrates’ guidance of “do no harm” is a good minimum threshold for any proposed improvement. Other criteria that would help control risk are those that can be used to assess the compatibility of user community astrodynamics algorithms with those used operationally by the JSpOC (i.e., what conditions should be met for algorithms to be considered “compatible,” such as, to within how many meters must an orbit propagation agree?).
The JMS schedule shows that AFSPC is working hard to meet a 2014 deadline to transition off the aging JSpOC SPADOC mainframe computer. This short time line makes it undesirable to incur the schedule risks of major astrodynamics algorithm changeouts during this 2-year period. During this time a strategic vision can be developed as to what new algorithms AFSPC wants to evaluate and cost estimates performed on impacts to the legacy community. After the 2014 deadline has been met, AFSPC should be able to start taking advantage of the JMS service-oriented architecture to implement upgrades.
It is also important to be able to balance the benefit of improvement versus the cost of implementation not only in JMS, but also for the user community as well. For example, in the 1980s AFSPC had developed an improvement to the workhorse general perturbations ephemeris routine SGP4. After rigorous testing, it was determined that this new routine, called SGP8, could improve prediction accuracy by about 10 percent for many satellites. However, because the new algorithm was not compatible with SGP4 users’ software and because format changes were required in the standard two-line element set products delivered to users, it was decided that the cost outweighed the benefit, and the algorithm was not implemented.
Limited funding is available to make changes in field units to match JMS changes. Modern Space Surveillance Network sites cost the Electronic Systems Center (ESC) in the range of $5 million to $10 million in sustainment expense each year. This cost does not include all the other operation and maintenance costs incurred for such things as generation of power to run the radar. More than 10 sites that are part of the Space Surveillance Network require compatibility with the JSpOC. Operation and maintenance funding does not allow for more than just running the sites. Improvements come only at long time intervals in a special maintenance category (called the System Lifetime Extension Program) that allows for such things as replacement of obsolete computer hardware. If the Space Surveillance Network sensors are to be impacted by JMS changes, then additional funding will be required to allow them to remain compatible.
In addition, outside agencies will require additional funding to modify their systems to accommodate changes. It is difficult to estimate the cost of such changes without a detailed knowledge of the modifications required and
the structure of the site/system to be modified. Based on past government software modification costs, each site/user system could incur costs of $2 million or more in additional funding to enable compatibility with each JMS-related upgrade that necessitates major changes to site/user software. Smaller changes would of course be less costly. Careful selection by JMS of the legacy user interface/algorithm modifications incurred by proposed JMS changes could mitigate this expense.