By Joseph Borson —
In early 2010, the White House quietly canceled the long-troubled National Polar-orbiting Operational Environmental Satellite System (NPOESS). The project would have consolidated almost all new U.S. polar orbit weather satellites into a single administrative structure, jointly run by the Department of Defense (DoD), NASA, and the National Oceanic and Atmospheric Administration (NOAA). These satellites, because of their unique orbital geometries, are critical for long-term climate change modeling (as opposed to hour-to-hour weather forecasts a la “weather.com”), and the original hope was that sharing satellite assets among different interested agencies would allow for efficiencies of scale, the sharing of technical expertise, and ultimately, better information.
It didn’t work out that way. Instead, the program’s budget ballooned from $6.5B in 2002 to nearly $14B in 2010, and the project suffered technical delays in satellite integration and imaging instrument installation – in other words, the satellites were blind. These types of issues are far from uncommon in satellite design programs. But, in the assessment of House Science Committee Subcommittee on Investigations and Oversight Chairman Brad Miller (D-NC), the program was ultimately felled because it was “crippled by a management structure that delayed decisions at critical moments.” The three agencies could not harmonize their distinct operational requirements, and ultimately, tried to create a project that could serve everyone – and a result, ended up serving no one. The remnants of the NPOESS project will ultimately be devolved back to the constituent agencies, with the DoD and NOAA each running a separate set of satellites. Data will hopefully be still be shared, but the dream of a single U.S. climate satellite system appears to be dead.
But is that such a bad thing?
From the perspective of modeling information, the answer is unequivocally yes. U.S. agencies had been depending on NPOESS as a source for critical forecasting information for the 2010s, and as old polar orbiting satellite systems wear out, there will be a gap when there is really nothing that can serve as a capable replacement. But without good data, there is no way to create the effective climate models necessary to determine the potential effectiveness of climate change policy. And as policymakers continue to recognize climate change as an increasingly important threat to national security, and one worthy of a significant regulatory response, the idea that they would implement policies while blinding themselves to the data necessary to inform that policy-making process is inexcusably short-sighted. The U.S. plans to try to mitigate damages by keeping satellites in orbit beyond their planned life, and by sharing data with international partners, but stop-gap responses simply cannot substitute for an effective space infrastructure.
But from the point of view of shared military-civilian scientific programs, NPOESS may serve as a practical reminder of the benefits – and limitations – of cross agency cooperation. NPOESS ultimately failed because incompatible institutions and institutional needs could not be grafted together in the interest of cost savings. This is not unique to satellite constellations; back in the 1970s, the DoD insisted that the space shuttle be designed so it could service military payloads, which it did regularly throughout the 1980s. But this forced design compromises in the shuttle itself which made it unwieldy and expensive, and ultimately helped to make it a far less economical – and useful – vehicle than its proponents had hoped.
This is not to say that all military-civilian cooperative efforts are doomed to fail. Indeed, one of the best examples of success is on your screen; the internet’s history as a joint project designed to facilitate university communication (and post-nuclear war communication) is well known. The earliest satellites of the 1950s were also the products of civilian-military cooperation – which eventually led to the development of GPS, communication, and (perhaps ironically, given the current history) weather satellites. Failure is certainly not pre-ordained.
So what policy lessons can we take for structuring future civilian-military scientific programs?
- Focus on differences, not just commonalities: In order for a program to even be considered dual-use, there must clearly be significant common elements. These are seductive; it’s easy to say, as with NPOESS, that the idea of avoiding duplicate satellite systems is a good one, and that combining forces on developing imaging technologies can only lead to positive effects. And that’s not inherently wrong. But policymakers need to look at the institutional structures that are underlying these projects. Do different agencies need the same type of data, or do they require customization? If there are customization needs, does this require new types of instruments? Are the management structures between the agencies compatible? Is there a “binding arbiter” who can create enforceable compromises? If so, would these compromises best suit the scientific and national security needs of the United States? If the answer is no, a joint project may not be appropriate.
- Sharing information does not require joint programs: Just because agencies do not jointly manage projects does not mean they cannot benefit from the work of their peers. Having one institution manage a program – but then share results with its sister agencies – is often times the best way of maximizing the nation’s overall information resources. A textbook example is earthquake seismographs. The U.S. military originally put in place a global network of seismograph stations to monitor underground nuclear testing. But this system also provided an incredibly important library of geological information, and revolutionized how we look at the Earth’s structure. We can use this as a model going forward – emphasizing not the management of dual use programs, but practical results that come from sharing information with multiple parties. Is the information as useful as it would be if it was created specifically for an agency? Probably not. But there is a much better chance that the program will get off the ground – and information that is 95% useful is far, far better than climate information that comes from a grounded satellite with a crippled eye.
- Implications for inter-agency cooperation: Problems of institutional cooperation are certainly not limited to scientific research; indeed, the nation’s spy satellites were originally consolidated in one agency (the National Reconnaissance Office) specifically to avoid the problems that felled NPOESS. But as the intelligence community discusses the best ways to cooperate amongst itself, it is important to remember that institutional sharing does not equal institutional projects, and that just because agencies have similar overall objectives, does not mean that their organizational policies are conducive to joint management. This may be – and probably is – a sign that institutional reform is necessary. But if so, it is absolutely critical to avoid trying to do joint-projects before reforming the agencies themselves; otherwise, the community risks failed projects, gaps in information management, and continued black marks in the eyes of Congress and other key stakeholders.
NPOESS’ legacy will live on in its successor satellites – and hopefully, as a reminder of how civilian and military agencies should – and should not – cooperate on joint projects.
Image courtesy of the U.S. Air Force