Conclusions and Recommendations
In this chapter we attempt to unravel the space physics paradox by reviewing the findings from our earlier chapters and revisiting the big science/little science controversy. We have assembled our findings in a manner that naturally leads to four major conclusions and four important recommendations that address the essence of the space physics paradox. We have relied more on the trends derived from our data than on absolute values for any given epoch. Our conclusions therefore reflect these trends, and our recommendations seek to change them—a change that can be made with no increase in overall funding for space physics.
Other reports have touched on these matters but in different contexts. For example, the Lederman report  delivers a disturbing anecdotal survey on the health of university research. The National Aeronautics and Space Administration (NASA) Advisory Council report  recognizes a decreasing number of research opportunities, increasing time scales for research projects, and the scientific community's need for a variety of research opportunities of different sizes. The Committee on Space Policy , in its recommendations to then-president-elect George Bush, discusses the need for a balanced program consisting of a stably funded base program supplemented by large, long-term projects. Such general reports complement the more specific case presented in this report for the field of space physics.
THE REALITY BEHIND THE PARADOX
Before discussing our conclusions, let us reexamine the paradox itself. On the basis of individual case studies, the data base assembled for this report, and
anecdotal evidence from colleagues, the committee concludes that the problem underlying this paradox is real—that despite substantial funding increases in space physics over the past 15 years, the conduct of research has become less effective, leading to increased levels of dissatisfaction in the research community. However, by asking questions about where the money has gone, why inefficiencies have developed, and who is feeling the dissatisfaction most keenly, this seemingly paradoxical situation can be explained.
Chapter 3 showed that overall research funding, as well as funding for space physics research, have increased at a pace well beyond inflation and now represent a larger share of the gross national product than they did 15 years ago. Furthermore, Chapter 4 showed that the size of the space physics community has grown at a similar rate. On the other hand, Chapter 5 established that in the core program the percentage of proposals funded and the funding per grant have generally decreased over this time, concluding that the base program has not kept pace with either the increasing size of the field or the general funding increase. Thus, even though total funding has increased at a rate similar to the growth of the research community, individual "small science" researchers must now write significantly more proposals to support their work than they did a decade ago. Chapter 5 also showed how increasing university overhead rates are compounding these problems. Finally, Chapter 6 discussed the changes that have occurred over the past two decades in the selection, management, and implementation of space physics research projects. The data show that these projects have become larger, more complex, and more expensive, which suggests greater opportunities for the research community. However, other findings are more sobering: launch frequencies (and total experiments deployed) have decreased; project implementation times have risen across the board, dramatically in some cases; heavy documentation requirements have been imposed; and projects increasingly require individual, new-start approval from a strapped U.S. Congress. As overall funding levels increased, with more dollars targeted for large projects, many of these changes were unavoidable. Some even seemed reasonable and necessary to maintain an appropriate system of checks and balances. However, the net effect has been the establishment of a system that causes major implementation delays; disproportionate study, planning, selling, documentation, and administrative activities; inadequate funding profiles for planned programs; and a less effective core research program. All of these findings are consistent with the increased levels of frustration sensed through discussions with colleagues throughout the space physics community.
REVISITING THE BIG SCIENCE/LITTLE SCIENCE ISSUE
We examined the ongoing big science/little science controversy in Chapter 2, described the general characteristics of "big" and "little" science, and reviewed the debate concerning the balance between the two. As we saw, there is
a tendency for science to grow, and to grow rapidly. As a consequence, the perceived size of any science effort will depend on when it is developed within its respective subfield. Similarly, the relative size of a scientific project will vary from subfield to subfield, as well as from agency to agency. Despite these changing perspectives, it generally is possible to distinguish between big and little science at a given point in time, in a particular subfield, and within a specific funding agency.
When we do this in space physics we find that both big and little projects have been used to advance the field to its present state of knowledge. Chapters 4 and 6 illustrated how the field of space physics has evolved from a small group of pioneering researchers probing the edge of space with balloons and rockets to a community of several thousand researchers using state-of-the-art tools to study the space environment from the earth to the stars. Chapters 2 and 6 also showed that both large and small projects have been used together from the earliest days of what we now call space physics. When things go well, little science supports big science through the results of its research and discoveries, and often itself evolves into big science endeavors. In a complementary fashion, big science provides platforms for larger and more complex experiments, and often supports little science directly by providing experimental opportunities for many additional researchers and groups.
At one time a strong synergistic relationship existed whereby everyone seemed to benefit. However, the extensive experiences of the committee members, confirmed through discussions with colleagues, and substantiated by the data assembled for this report, indicate that this synergism has broken down. In struggling to explain what went wrong, we have found what we feel are important clues for understanding the paradox underlying this study. The consequences of our findings are embodied in four major conclusions presented below.
Conclusion No. 1: The effectiveness of the base-funded space physics research program has decreased over the past decade.
We saw in Chapter 4 that the size of the space physics research community has increased at a rate roughly commensurate with the general increase in research funding described in Chapter 3. However, Chapter 5 showed that the average grant size in the base-funded program (the source of support for most small science) has decreased during this time. This was brought about by an effort to fund a growing number of proposals from a budget that, while increasing slowly, has not kept pace with demand. In other words, the base research program has not participated fully in the overall funding increase. We estimate (Chapter 5) that researchers must now submit two to four proposals per year to remain funded, even more if graduate students, a research group, and instrument
development staff are to be supported. This contributes to the greatly decreased efficiency of the present core program. Much more effort is now expended per dollar on writing and reviewing proposals and on contracting for the research being done. Increased university overhead costs further exacerbate this inefficiency.
In an attempt to quantify this phenomenon, we estimated in Chapter 5 that the dollar value of the effort expended in writing, reviewing, and granting funds in the core program can reach up to 50 percent of the amount being awarded. From any perspective this is an unreasonably high (and generally overlooked) cost burden for an already stressed core research program.
Small science, carried out by an intellectually diverse, flexible community of independent investigators, provides unique capabilities for performing certain kinds of research. Because of this, it is our conviction that a strong, effective, base-funded research program is essential to the health of the field.
Conclusion No. 2: Factors such as planning, marketing, the funding process, and project management have become as responsible for the increased delays, costs, and frustration levels in space physics as technical complications related to increasing project size and complexity.
Chapter 6 showed that many space physics programs, both space-based and ground-based, exhibit the same trend of increasing time from conception to implementation. This trend is most pronounced in what we characterize as big programs, but similar problems are creeping into the smaller programs as well. For example, satellite mission implementation times have increased from two or three years in the early 1960s to the present value of 10 to 15 years. We also looked at NASA's Explorer program, where implementation times have become so great (Figures 6.3 and 6.4) as to undermine the original intentions of the process itself—namely, to do high-priority science in a timely manner.
Chapter 6 and Appendix B show that accompanying these increased implementation times has been a major increase in project management functions such as study, planning, review, and selling activities. These activities occur well before the start of a program and continue far into the implementation phase. They are time consuming, expensive, and often do not contribute much to the science being pursued. In some cases these efforts are for naught, and the planning never comes to fruition (see Appendix B). In too many other cases , these efforts have extended through such long study and implementation phases that a sense of disillusionment arises.
The management system further imposes documentation requirements that represent a substantial part of the experimental team's effort. The extent to which this requirement has been imposed is seen as excessive by the space physics community. For example, in many cases a separate manager is assigned the responsibility of collecting each set of required documentation.
Using examples from solar observatories, rocketry, and ballooning, Chapter 6 goes on to show that increased size and complexity are not the only, or sometimes even the major, factors in the increased implementation times and costs of research projects. For example, balloon and rocket program experience has shown that success rates and scientific productivity can be maintained despite significant growth in the size and complexity of instruments. By contrast, ground-based solar observatories show a major increase in implementation times, even though their size and complexity have remained comparable over the past two decades. In the latter example the increased implementation times are due to the extra management activities described above, combined with the vagaries of incremental funding that rarely matches the planned profile.
This management structure may be a natural result of the trend toward large programs. Because of the huge investment of resources in these large projects, government and other managers feel a responsibility to closely monitor every aspect of their progress. Unfortunately, the indiscriminate application of this management system to programs and projects of all sizes has reduced the effectiveness and increased the cost of the overall space physics research program.
Conclusion No. 3: The long-term trend that has led to an ever-increasing reliance on large programs has decreased the productivity of space physics research.
The time period that saw a growing reliance on large science projects also witnessed several disturbing trends. As described above, there has been a steady increase in the implementation times of space physics projects, both ground and space based. This has been accompanied by a steadily expanding effort by the research community on planning, study, and selling activities.
In parallel with these trends, the average grant size in the base-funded program and its constant-dollar buying power have decreased relative to inflation, despite the fact that overall funding for space physics has increased markedly over the past decade. We conclude by implication that the bulk of the funding increase experienced in space physics research over the years has gone to large programs. Because of their broad goals and national visibility, these projects have been easier for funding agencies to sell than the base research program.
However, the complex and ambitious character of big science projects also has its downside, including frequent cost overruns. The infrequency of largescale activities produces pressure to add on additional experiments that may not be essential to the primary goal of the project. The addition of these lower-priority components contributes to rising costs.
Unfortunately, small budget perturbations in big programs can have major effects on small programs. The OTA report  aptly illustrates this concern for the broader science community by considering cost projections that include the possible effects of four megaprojects: the Human Genome Mapping Project, the
now-cancelled Superconducting Supercollider, the Earth Observing Satellite, and the Space Station. It shows that in a scenario in which the total science budget is held to a specified rate of increase, constant-year dollar funding for the nation's science base will stay level or decrease through the 1990s, squeezed out by the big projects.
Big projects are also much more vulnerable to shifting political winds and competing science priorities. Tying science goals to new-start approval decisions with incremental funding allotments introduces a high level of uncertainty and risk. It has also led to a steady decrease in experimental opportunities (including balloon, rocket, and satellite flights and instruments) and a space physics community too often diverted from direct scientific research into peripheral management activities, or writing multiple proposals for the scarce dollars left in the base-funded program. A disturbing, but telling, side effect of the processes described above is that a steadily decreasing percentage of experimentalists are entering the space physics field, as noted in Chapter 4.
We believe that these trends must be halted and reversed in order to restore the health, and safeguard the future, of the field. We also believe that the space physics story may contain lessons of value to the broader academic community.
Conclusion No. 4: The funding agencies and the space physics community have not clearly articulated priorities and developed strategies for achieving them, despite the fact that the rapid growth of the field has exceeded available resources.
The number of researchers in space physics has grown considerably over the years (Chapter 4). The growth has been accompanied by a marked increase in the number and complexity of new research problems proposed within the field. Lacking clear guidance from a set of ranked priorities, the funding agencies have absorbed into their strategic plans more ideas and programs than could be implemented within the bounds of available, or realistically foreseeable, resources. Many of these programs were then maintained in readiness, awaiting the availability of formal approval and funding in the face of competing national priorities. Some of these projects were never started (Appendix B), or were canceled in midterm (Chapter 6), wasting resources and failing to achieve the scientific goals that drove them. Even those that do see completion now take longer and cost more as a result of this process (Chapter 6). The funds consumed in maintaining programs in readiness do not represent an effective use of resources and often impact the core science program as well.
Based on the four major conclusions presented above, we have developed four interrelated recommendations. It is worth repeating the caveat expressed in
Chapter 5 that this report examines trends in the conduct, not the content or quality, of space physics research. Nevertheless, we believe that implementation of the recommendations in this report could greatly increase the amount of productive research accomplished per dollar spent in the space physics community and could significantly reduce the level of frustration without increased funding levels in the overall space physics research budget.
Recommendation No. 1: The scientific community and the funding agencies must work together to increase the proportionate size and stability of the base-funded research program.
As has been discussed, a productive space physics program cannot survive on large new-start projects alone. An active and synergistic program of small projects is also needed to incubate new ideas. To support this family of small projects, there must be a stable and effective base-funded research program. Furthermore, a revitalized core program must adopt procedures to decrease present inefficiencies (Chapter 5). With a larger, more stable core program, the funding agencies can increase grant sizes and durations, enabling researchers to focus more on science and less on funding. Other improvements in the funding process, such as requiring shorter proposals, providing a faster, more efficient review process, and delegating greater authority to the principal investigator where possible, are discussed in more detail in our last recommendation.
Recommendation No. 2: The funding agencies should ensure the availability of many more experimental opportunities by shifting the balance toward smaller programs, even if this necessitates a reduction in the number of future large programs.
The current frequency of experimental opportunities is insufficient to sustain space research into the next generation. As explained in Chapter 2, some scientific investigations can only be done via large science initiatives. In fact, we have shown in Chapters 2 and 6 that a strong synergism existed at one time between large and small space physics programs. However, the future of space physics requires frequent access to new research opportunities and the accompanying development of new scientists capable of carrying out the missions of the future. These goals have been more efficiently achieved through small science programs.
In one survey of the space physics community by NASA's Space Physics Division , researchers were asked to characterize their desires for the future of space physics with regard to small and large missions. Although these results must be qualified in light of a low response rate, over 90 percent of respondents expressed a preference for more frequent access to space through small mis-
sions, even if it meant a reduction in the number of large missions. Our recommendation concurs with this view.
Recommendation No. 3: In anticipation of an era of limited resources, the space physics community must establish realistic priorities across the full spectrum of its scientific interests, encompassing both large-and small-scale activities.
Unless the scientific community itself is willing to make difficult choices and set priorities at the outset, programmatic decisions will ultimately be made on the basis of considerations other than a rational assessment of the value of the program to the nation's scientific progress. The needs and vitality of the discipline as a whole, and not necessarily equity among subdisciplines, are of paramount importance in setting such priorities. Thus, these overall priorities must be considered before it becomes productive to prioritize specific programs and missions. Long-term scientific goals should not be altered lightly or set aside before they are achieved. Scientific priorities should change only in response to changing scientific perspectives. Ongoing projects initiated in response to established scientific priorities should be insulated as much as possible from the effects of short-term fluctuations in funding.
Prioritization must not only cross subdisciplines within a field but also include an assessment of the balance between ''big'' and "little" science. In particular, it must consider the unique value of a large project relative to the promise of the ongoing base-funded program. A clear assessment of this delicate, yet important, balance must be made by the research community at the outset.
Prioritization is always a wrenching process, since it necessarily involves postponing or eliminating the pursuit of some interesting ideas. It is beyond the scope of this report to recommend a specific process for making these hard choices; however, the space physics community might be able to learn from other academic disciplines that have been forced to undertake similar priority-setting exercises.
Recommendation No. 4: The management and implementation processes for the space physics research program should be streamlined.
Management and implementation processes must be tailored to the size of a given program: big science management techniques should not be applied to little science programs. Politicians, funders, and managers need to acknowledge that their understandable desire for accountability and program control exacts a price in inefficiency, delay, and, occasionally, failure to achieve scientific goals.
Oversight and reporting requirements should be reduced in many instances, even at the expense of assuming a somewhat greater risk. Risk is an everpresent and essential element of the scientific enterprise and should be accepted
at levels in keeping with program size. Study, planning, and selling activities should be reduced and implementation times should be shortened to provide increased experimental opportunities, allow for the timely pursuit of research problems, and ensure the training of students. Proposal reviews also should be streamlined, particularly within the core program.
Many of these steps will be easier to accomplish if a set of priorities is developed by the scientific community. Such priorities will also aid the funding agencies in their programmatic decisionmaking. For their part, funding agencies must work with the scientific community to streamline management procedures, consult with the scientific community as they make programmatic decisions, and clearly convey, and then fulfill, their level of commitment throughout the development of a program.
We feel that major progress can be made by recognizing and relying on the powerful self-interest of researchers to succeed. For example, the continued success of the rocket and balloon programs, despite their increased size and complexity, can be traced in part to the fact that management of these programs has been left in the hands of the principal investigators (Chapter 6). We believe that delegation of greater authority to principal investigators will generally lower the direct and indirect costs of oversight and reporting while improving success rates and scientific productivity.
The four recommendations outlined above are highly interrelated. Streamlined management processes will further boost the productivity of a stabilized core program. Priority setting will enable the few most critical big science projects to be pursued without jeopardizing ongoing research. Taken together, we believe these recommendations provide a blueprint for a stronger and more productive space physics research community.