Program Planning, Structure, and Organization
Many of the present programs that started during the late 1980s are now in or are nearing a synthesis stage, and are projected to end in the next few years. Consequently, this is an appropriate time to review and critique the planning, structure, and organization of the existing programs for the benefit of future programs. To identify systematic changes that could improve efficiency, enhance collegiality, and increase the likelihood that future programs will achieve the scientific goals set out for them, the committee developed a study approach (Box 6-1) that emphasized a generic review of completed and ongoing programs.
Nature Of Major Oceanographic Programs: A Comparison
The spectrum of past and present major oceanographic programs can be subdivided into three groups, based on their origin and histories: (1) current programs with major NSF involvement; (2) programs initiated as part of IDOE; and (3) programs managed through the Office of Naval Research. To a large degree, these programs were developed to address different scales and types of scientific questions. Thus, they tend to represent a spectrum of structures. Members of each group have inherent strengths and weaknesses that may need to be considered before establishing models for future programs, thus each group was examined in an effort to provide insight into way future programs may initiated and/or structured.
Box 6-1 Task Group 4 Study Approach
Task 4) The committee will recommend how future major oceanographic programs should be planned, structured, and organized.
Question 4: Which existing program structures and organizations worked well and should be used as a pattern for future programs?
The Present Programs Involving NSF
To better understand the variety of approaches and mechanisms involved in program initiation, the initial phases of several existing or former programs were examined using program publications and responses to the questionnaires (Appendices C, D, and E). The origin and evolution of the programs varied in detail, but many common traits are apparent.
TOGA: Two converging workshops were held to start TOGA. One workshop was conducted at Princeton under the auspices of the NRC Climate Research Committee (CRC). This meeting led to the 1983 NRC report El Niño and the
Southern Oscillation—A Scientific Plan. An international workshop led to the formation of the WCRP. Both workshops recommended a program to study El Niño and the Southern Oscillation. The CRC was the founding committee in the United States since it had liaison duties to the WCRP (which was planning international TOGA).
ODP: The Ocean Drilling Program was initiated to continue the ocean sediment drilling efforts of U.S. scientists started with the Deep-Sea Drilling Project (DSDP), which ran from 1968 to 1983. In 1983 the Joint Oceanographic Institutions, Inc., (JOI) Board of Governors nominated an interim U.S. Science Advisory Committee (USSAC) to serve as a national committee to coordinate drilling-related activities. USSAC met in 1983 and prepared a proposal for a U.S. science program in this area. ODP was initiated in 1984.
WOCE: A few key individuals proposed a program to the Committee on Climate Changes and the Ocean (CCCO) and the WCRP. In the United States a panel was established in the NRC that reported to Ocean Sciences Board (the predecessor of today's Ocean Studies Board). In 1983 that panel organized an international workshop under the aegis of the NRC to consider the scientific feasibility and possible approaches of a research program to study global ocean circulation. It was agreed that such a program was sensible, and planning for WOCE was begun under the international sponsorship of the CCCO and the Joint Scientific Committee (JSC) of the WCRP.
JGOFS: US JGOFS began as Global Ocean Flux Study (GOFS) in 1984, and the international JGOFS program started three years later. A brief history of the evolution of this project was published by McCarthy in 1992. Three initially separate committee projects converged during the early 1980s to make projects like GOFS and ultimately JGOFS possible: (1) the NRC Ocean Science Board investigated the feasibility of a program to conduct long-term studies of the biological and chemical dynamics in the ocean on basin-wide and global scales; (2) the NSF Advisory Committee for the Ocean Science Program developed a long-range plan titled Emergence of a United Ocean Sciences; and (3) a separate NRC committee identified initial priorities for International Geosphere-Biosphere Programme (IGBP). With support from NSF, NASA, ONR, and NOAA, the NRC convened a workshop in Woods Hole, Massachusetts., in 1984. As a result of the workshop, the GOFS Scientific Steering Committee was formed in 1985.
RIDGE: The NRC funded a workshop at Salishan Lodge in April 1987 on ''The Mid-Ocean Ridge—A Dynamic Global System" and a scientific steering committee was formed shortly thereafter This workshop was followed by three workshops in 1988 on specific aspects of the proposed RIDGE Program. These topics were Mapping and Sampling; Sea-Going Experiments; and Theoretical,
Experimental, and Analytical Approaches. Reports from five RIDGE Working Groups and a draft program outline were published in 1989.
CoOP: In 1985, the first Advisory Committee for the Ocean Science Program developed a long-range plan entitled Emergence of a United Ocean Sciences, which included discussion of the value of coastal research program. An open meeting was held after an American Geophysical Union (AGU) Ocean Sciences meeting in 1987, followed by a series of smaller workshops across the country to discuss program goals, scope, and objectives. As the program evolved it became interdisciplinary, and the name was changed from CoPO (Coastal Physical Oceanography) to CoOP (Coastal Ocean Processes).
GLOBEC: Emergence of a United Ocean Sciences also described a program with many of the elements present in today's GLOBEC program. The origin of U.S. GLOBEC was associated with a meeting sponsored by NSF, NOAA, and ONR in 1988, which led to the formation of the Scientific Steering Committee in 1989.
CLIVAR: In 1991, while TOGA and WOCE were still in progress, the Joint Scientific Committee (JSC) for the WCRP called on a group of experts to consider possible future directions for climate research. The deliberations of this study group were published in 1992 under the title CLIVAR—A Study of Climate Variability and Predictability . The JSC formally decided in 1993 to undertake CLIVAR as a major new activity in the WCRP and established a Scientific Steering Group (SSC). CLIVAR was initially divided into three programs: (1) CLIVAR-GOALS (the ocean component of CLIVAR) is a study of seasonal-to-interannual climate variability and predictability of the global ocean-atmosphere-land system; (2) CLIVAR-DecCen is to address decadal to centennial climate variability; and (3) CLIVAR-ACC, which concentrates on modeling and detection of anthropogenic climate change. The NRC organized a workshop for GOALS in 1992 to produce a science plan (NRC, 1994b). In the United States, CLIVAR planning is being coordinated by two NRC panels, one for GOALS and another for decadal to centennial climate variability and predictability (CLIVAR-DecCen) and anthropogenic climate change (CLIVAR-ACC).
The scientific questions behind the present set of major oceanographic programs came forward largely from the oceanographic community. In many cases, the scientific challenges that became the focus of these programs began with a few individuals, often colleagues with a history of collaboration. These kinds of collaborations are sometimes facilitated in the large oceanographic institutions where researchers have easy access to colleagues in a number of relevant fields and a potent research infrastructure that includes a variety of laboratory facilities and computer services. Thus, many of the ongoing programs were nurtured by early associations with large institutions.
Public forums at national meetings were often used in an effort to solicit broader community input and discussion. Although these were useful for information dissemination, they were not usually an optimum approach for reaching a community consensus. Conversely, planning workshops with smaller numbers of attendees and more focused agendas appear to have been the most successful mechanism for actually getting major ocean programs started. Workshops planned through the National Research Council have played an important role in this regard (e.g., WOCE, TOGA, JGOFS, and RIDGE). JOI has also organized workshops that led to major oceanographic programs (e.g., GLOBEC). Workshops organized and sponsored by professional societies (e.g., AGU-Chapman, GSA-Penrose) could also be used.
In most cases a scientific steering committee (SSC) was formed at the initial workshop, and it guided the ensuing planning. The planning workshops and resulting reports were a successful approach for incorporating a broad range of scientists from the community into the planning process. This approach seems to have worked well and should be used for future major oceanographic programs. Workshops are an effective way to start new programs if they:
- actively recruit participants to include a diverse and broad-based spectrum of the community; thus ensuring that they are (and are perceived to be) open;
- develop a product of high scientific credibility and broad based support in the community and in the funding agencies for program goals; and
- solicit international attention and coordinate with parallel international activities.
In practice, the success of an initiative for a new major oceanographic program has relied on the vision and energy of a few key individuals. Efforts must be made to allow motivated individuals to act as advocates for ideas while allowing for ever-increasing involvement by the broader community.
International Decade of Ocean Exploration (IDOE) Programs
As discussed in Chapter 2, many of the existing programs can trace their intellectual roots back to programs developed and executed as part of the IDOE. To better understand how major federal initiatives can influence the development of programs, the committee reviewed the history of IDOE with the objective of drawing lessons for the future. Lambert (in press) also provided a summary of the history of major oceanographic programs at NSF.
During IDOE, ideas for projects originated entirely from individuals in the scientific community. For example, Henry Stommel provided initial inspiration for GEOSECS when he recognized that geochemical tracers had great potential for providing understanding of deep-ocean circulation and mixing. Ultimate
scientific direction for GEOSECS was provided by a 12-person Scientific Advisory Council, and a project director supervised coordination of the shipboard operations. Maurice Ewing played an early role in MANOP by encouraging a conference held at Arden House in 1972. A steering committee was formed at that meeting that resulted in the first version of the MANOP. However, the project had problems finding a clear focus and was eventually restructured at a meeting at Batelle Northwest in Seattle, Washington in 1976. Most other IDOE projects were similar to GEOSECS and MANOP in that one or two motivated individuals provided leadership during the formative phases. There was generally an extended iteration period with the IDOE program office that involved trips to Washington, D.C. requests for planning funds, and initial planning meetings. Omnibus proposals were prepared, which were reviewed by panels separate from the core NSF programs. In most cases open community meetings were not held to shape the science and there were no community-wide calls for proposals. When they began, most IDOE projects had no clearly identified end and went through several proposal cycles. Most tended to put more energy into field programs than into synthesis and modeling.
The IDOE approach had strengths and weaknesses. On the positive side, IDOE programs typically involved small groups of self-selected scientists; thus, the administrative costs were small and collegiality within the project was largely quite good. The programs usually held annual principal investigator meetings, and internal communication was very good. For the most part the resulting projects were an efficient, direct approach to specific problems (NRC, 1979). However, once formed, it was difficult for the programs to incorporate new investigators. In addition, some of these IDOE projects were more successful than others in reaching their goals. In many cases the projects were open-ended and went through several proposal cycles.
As previously mentioned, the NRC, at the request of NSF, reviewed the initiatives conducted as part of IDOE and made a number of recommendations regarding a potential extension of that effort (NRC, 1979). In addition, a number of recommendations regarding program organization and management appear in the report that should be considered by future programs:
- A variety of approaches to project development should be encouraged. It is particularly important that the process be open to all scientists who are qualified and interested in participating.
- For major projects of long duration, widespread participation in their development should be encouraged through early notification of planning meetings. Depending on the nature of the proposed research, scientists and engineers from a variety of institutions and representatives of government agencies should be included.
- The process of peer review should continue to emphasize scientific quality and significance and the qualifications of investigators. Review panels should
- Responsibilities for elements of future program should be allocated among NSF program managers on a flexible basis, with projects concerned with related scientific, logistic, and operational problems grouped under a common program manager.
- At the operating level, projects should continue to be managed under arrangements developed by participating scientists. These arrangements should be highly responsive to scientific needs and be able to accommodate new ideas as they arise.
- Projects should be structured into logical work and time segments, and the review schedule should take advantage of natural development plateaus. Project duration should be made clear by the early establishment of well-defined termination points.
seek opportunities to combine projects or elements thereof where there is significant overlap.
Office of Naval Research Programs
During the 1940s and 1950s, the Department of the Navy was the predominant sponsor of oceanographic research. The Office of Naval Research (ONR) continues to play an important and complementary role to NSF in supporting ocean sciences. The spectrum of scientific projects funded by ONR is different from NSF, in part, because ONR places greater emphasis on intermediate-sized programs than does NSF. These programs typically involves less than 30 principal investigators, and when present, the steering committee generally includes only 2 or 3 individuals. The initiation of intermediate-size programs at ONR typically involves an iteration between a specific program manager, other program managers within ONR, and scientists in the field. The program manager finds sound ideas that will be consistent with ONR objectives and consults with experts in the field about scientific approaches. This relationship between the ONR program managers and principal investigators in the community works well for a select group of investigators. Once a topic is chosen the program manager puts together a package of investigators and proposals. The scientific team is selected by ONR. There is usually an announcement for unsolicited proposals for specific initiatives.
A typical ONR initiative has a focused scientific objective and lasts about five years. A program schedule might include a planning year, two field years, and a period for data analysis. As a result new initiatives are in the planning stage as existing initiatives are winding down. The scientific support is continuously recycled to address the scientific questions of greatest importance to ONR and the Navy.
ONR programs contain a few features that make them markedly different from major programs currently supported in large part by NSF/OCE. ONR programs tend to be very focused and therefore somewhat insular once under
way. This structure appears to work well in settings where the scope of the program needs to remain fixed over the duration of the project. Such a closed structure would probably not be desirable for studies that require a more flexible scope and thus need greater interaction with the broader research community. Yet there are other features of ONR intermediate-size programs that should be considered for incorporation into new programs. For example, ONR programs have a well defined length and end; thus, ONR programs are continuously forming and dissolving so that they have a relatively short residence time and don't become stale. NSF/OCE and other sponsors and organizers of any incipient oceanographic programs should maintain the flexibility to consider a wide range of program structures before choosing one that best addresses the scientific challenge.
Future Program Initiation
The current set of major ocean programs have been uneven in the degree to which their scientific objectives and motivating hypotheses were well matched. In some cases worthwhile scientific objectives did not lead to a program that could realistically deliver what it set out to do. During the initial planning and organization of new major oceanographic programs, efforts should be made to ensure agreement between scientific objectives and motivating hypotheses given for funding. Recognizing that there are some areas where insufficient information exists to form focused hypotheses, programs of a more descriptive nature may be required. However, when possible, greater emphasis should be placed on developing focused, hypothesis-driven projects with specific and tractable objectives. It is important to ensure that scientifically sound ideas that may require major programs can come forward. The sponsoring agencies, especially NSF/OCE, should develop well defined procedures for initiating and selecting future major ocean programs. One approach, similar to what is done for Science and Technology Centers, would be to solicit preproposals from the entire community, with the criteria for judging these proposals clearly stated in an announcement of opportunity from NSF. Regardless of the specific approach taken, the final mechanisms should incorporate the recommendations put forward in the 1979 NRC report The Continuing Quest: Large-Scale Ocean Science for the Future and reaffirmed earlier in this chapter.
Once potential topics that might warrant large organized research efforts have been identified, a series of workshops to encourage participation from a broad range of relevant expertise should be held to help shape the scientific goals and the implementation plan. In addition, initial planning workshops for new major oceanographic programs should be administered by an independent group and structured to ensure participation by a broad range of scientists including those from large and small institutions, different disciplines, and minorities.
Structure Of Future Major Programs
Major oceanographic programs represent a significant commitment of human and fiscal resources, and the coordination of these efforts is a major undertaking in its own right. The effectiveness, and to some degree, the character of these major programs can be greatly influenced by the program's structure. For the purposes of this report, the structure of a program can be characterized by its administrative organization (e.g., existence of scientific steering committee, program office), its size and length (e.g., number of principal investigators, years of duration), and the nature of its national and international relationships. It is important that the structure employed encourage planning as a continuous process.
Role of Scientific Steering Committee
Major oceanographic programs tend to have a scientific steering committee (SSC) with an executive committee and a chair. The SSC members have averaged two-year terms, and for most programs there has been a large turnover resulting in broad community involvement. What the steering committees actually do and how the decision making is divided among the chair, executive committee, and steering committee varies greatly from program to program. In general, the role of the SSC has been to:
- prepare a science plan;
- prepare an implementation plan;
- design specific activities to accomplish those plans;
- give advice as to priority and relevance of submitted proposals to funding agencies;
- provide continuing creative leadership;
- initiate mechanisms for turnover of SSC members so that broad community involvement is attained; and
- evaluate the scientific program direction and make changes as required.
The SSC needs to be large enough to accomplish these tasks, maintain continuity, and also account for representation and participation from the community. Most programs also have a program office overseen by a project officer. The role of the program office is to implement the science activities and logistics.
Other management structures have been used successfully. IDOE programs usually had a project director and a small executive group. Funding was usually provided for a logistics assistant to take care of cruise preparation and arrangements for meetings, etc. In lieu of a steering committee in most cases, investigators got together at least once a year to make programmatic decisions. TOGA was managed by an NRC panel comprised mostly of investigators. The panel's
duties were similar to those of an SSC, yet were subject to the oversight of an NRC board. ONR projects are sufficiently small that a scientific steering committee sometimes is not used. The principal investigators assemble once or twice a year. One of the lead principal investigators is a Project Director who works closely with the relevant ONR program manager.
There is no single, ideal organizational structure that should be used for all programs. The structure of the program should be dictated by the complexity and nature of the scientific challenge it addresses. Each new program should involve the simplest, most open structure required to implement the science. The nature and support of the administration should reflect the size, complexity, and duration of the program.
There needs to be flexibility so that the size and length of the program is no greater than required to solve the scientific problem. The focus and goals of present major ocean programs has tended to be large-scale and far-reaching (Box 3-2), thus the resulting programs have tended to be global in view and of at least a 10-year duration.
The present NSF/OCE structure has made it difficult to get intermediate-size projects (as distinguished from major programs) funded, particularly ones that are interdisciplinary. It has also been difficult for multi-level proposals (observations, theory, and modeling) to succeed. One problem is that these interdisciplinary and multi-level projects may not be on the cutting-edge of all disciplines and levels. Federal sponsors should have the flexibility to support disciplinary and interdisciplinary projects that span a range of sizes. For example, there could be scientific problems that can be addressed with one or two field seasons by 5-10 investigators, without an SSC and in the normal duration of an NSF grant (5 years). These intermediate-size projects could be solicited, funded, and executed in a way that would ensure a regular turnover of new ideas and opportunities for different investigators. It is possible that some concepts suggested for consideration as new major programs may be more appropriately addressed by intermediate-size programs. Conversely, intermediate-sized programs may be seeds for future major programs. These intermediate-size programs will not work for all of the scientific challenges; however, it is important that adequate flexibility exist to ensure that a spectrum of well-honed, hypothesis-driven programs can be developed.
There has been no standard procedure for determining the end of the present set of major oceanographic programs. TOGA was the only major oceanographic program that had an end defined when it started and that was for a duration of 10 years. The rationale was that a 10-year data record was the minimum required
for describing ENSO events and studying their predictability. JGOFS and WOCE are scheduled to end in 2002. GLOBEC is a relatively new major oceanographic program and has no perceived end at present. RIDGE and ODP have no defined end.
There is a perception within some segments of the community that once the science and implementation plans of major ocean programs are set in motion there is inadequate flexibility, and that the programs tend to be too intellectually constraining. This is further complicated when programs have no specific end dates. All programs should have well defined milestones, including a clearly defined end. These milestones should reflect the minimum effort and time needed to adequately address the scientific challenge the program has defined. An iterative assessment and evaluation of scientific objectives and funding should be undertaken in a partnership of major ocean program leadership and agency management.
When the nature of the scientific challenge dictates that program duration extends beyond five years, mechanisms should be implemented to ensure that the program continues to meet its objectives and milestones. Specifically, these longer programs should be reviewed periodically by an independent panel of scientists, which would:
- examine the program's scientific objectives and evaluate the program's progress toward them;
- recommend changes to scientific goals and the implementation as needed, in acknowledgment of changing objective and funding constraints;
- identify opportunities for greater community involvement; and
- report to the community and sponsors regarding the health and vitality of the program.
Synergy Between Models and Observations
While major ocean programs have lead to closer communication between observationalists and modelers, future programs should have a vigorous synergy between them. For example, during design of field studies, there are numerous stages at which this synergy can be developed for predicting temporal and spatial variability. Likewise, models should take the latest field studies into account and use measured rates and fluxes as constraints. The results of this synergy should enable extrapolation and prediction of knowledge gained from field and model exercises to a larger range of scales and environments. Plans to address this synergy should be included in the science plan and proposal for future major oceanographic programs. Such interaction naturally would be expected as part of a synthesis, but it should be conducted throughout the program as well. Although the synthesis phase of the present generation of major ocean programs relies heavily on modeling and data assimilation, these components should be included
during the planning and all stages. Modelers and observationalists need to work together during all stages of program plan design and implementation.
Coordination with Long Lead Time and Mission Agencies, and Ship Scheduling
One of the strengths of the major programs has been their ability to direct a significant amount of talent and scientific interest toward a large and often high profile scientific challenge. Consequently, these programs have been able to secure support for basic research outside NSF. For example, although WOCE was initiated with NSF funds, it received support from a number of mission agencies. Although NSF provided a significant share of the funding for individual principal investigators, other resources included contributions of personnel, facilities, satellites, etc. (Fig. 4-1). JGOFS included significant NOAA, ONR, and NASA components as well. It is doubtful that an equal number of principal investigators working alone or in small groups could have addressed a problem of such significant scope without similar resources being made available to them by federal agencies.
In some instances the present programs wrote their implementation plans in anticipation of new remote sensing data being available. Delays in providing that data limited achievement of some of the initial goals. The opportunities to use remote sensing data are rapidly expanding as they provide global scale temporal context for programs. In return, the satellite data are dependent on high-quality in situ data, often provided by major oceanographic programs. Future programs will likely be more closely coordinated with the satellite platforms.
Trying to maximize the resources available, in such a consortium of sponsors, requires significant levels of cooperation and coordination. This coordination hinges on good continuous interaction among program managers. There must be a commitment by all agencies to meet the needs of the program objectives. There are recent examples where agencies failed to keep their commitments: the U.S. JGOFS Arabian Sea process study, where ONR made last minute adjustments in its support, and where the DOE support for the U.S. JGOFS global carbon survey was removed before the program was completed. DOE also prematurely terminated its support for carbon fluxes from the Ocean Margins Program. As another example, the U.S. WOCE North Atlantic experiment was planned with the NOAA Atlantic Climate Change Program, and NOAA did not keep its commitments. These disappointments are reminders that priorities are set differently in and for the different agencies, and this needs to be considered in the planning process.
By definition, the science goals of major oceanographic programs are comprehensive and often cross discipline boundaries, casting a net greater than the programmatic venue of a given funding agency and beyond the mission of a single agency. For the present major programs, interagency coordination has
been achieved with variable success. Attainment of the goals of interdisciplinary major programs has been jeopardized by the variable level of commitment by different agencies. Although regular interagency meetings would foster better coordination, planners and organizers of new major programs need to recognize that mission agency funding profiles are different and may be shorter than the time scale of a major ocean program. Therefore, major programs need to maximize the chance that agencies can maintain their funding commitment by working to identify research support so that it is consistent with mission agency priorities. When the scale and complexity of the program warrants, an interagency project office should be established. Other mechanisms, such as memoranda of understanding (MOU), should be used to ensure agency support throughout the program's lifetime.
Ship scheduling is another coordination challenge facing major oceanographic programs for implementation planning. The large number of ship days requested at specific times—often with particular capabilities and sometimes on multiple platforms requiring synchronized schedules—represents a major factor for the UNOLS scheduling process. Shifting of projects from originally proposed vessels, schedules, and ports effects both major ocean program and core investigators and their budgets. These scheduling decisions are frequently made without consideration of the costs to the scientific investigation, such as travel and shipping. When possible, major program ship scheduling should be completed at least one year prior to the field work. Contingency funds should be made available to mitigate the budgetary impact of late changes in ship scheduling on science.