4
Other Facility Management Models
The workshop session continued with presentations from facility managers of the International Ocean Discovery Program (IODP), the National Ecological Observatory Network (NEON), and the National Aeronautics and Space Administration’s (NASA’s) Earth Observing System Data and Information System (EOSDIS). These speakers had been asked to describe their management and governance structures based on the same prompting questions provided to Detrick and Miller prior to the workshop (see Box 3.1). These presentations were followed by a panel of all of the facility managers that gave presentations. Moderated discussion prompted identification of commonalities and differences among the various management, governance, and decision models. The intent of the session was to provide workshop participants with a suite of management and decision models to consider when thinking about future seismological and geodetic facilities (to be discussed in a later session).
This chapter includes summaries of each of the three presentations given by facility managers, followed by a section describing the major themes discussed during the panel discussion.
MANAGEMENT OF THE INTERNATIONAL OCEAN DISCOVERY PROGRAM (IODP)
Bradford Clement, Texas A&M University
Bradford Clement is the director of Science Services for IODP located at Texas A&M University. The IODP is a multi-national program with about 26 countries that participate, fund, and coordinate program activities. The current program provides the science community with the only means to access sediments and rocks deep within the seafloor. It also installs deep seafloor observatories. IODP has evolved several times since its inception (see Box 4.1, presented by Clement) and has operated under multiple management structures. Clement described his involvement with the program under three different management models (under a contract, a multi-tiered contract as a subcontractor, and then later under a cooperative agreement).
Clement noted that today, IODP is a federation of affiliated drilling operations with three platforms supported by different agencies: (1) the National Science Foundation (NSF) supports a non-riser drilling platform (the JOIDES Resolution, which Clement manages); (2) a consortium of European countries supports mission-specific platforms; (3) and Japan provides riser operations using the drillship Chikyu. The JOIDES Resolution has undertaken the majority of the expeditions for much of the program’s history. Because the three platforms can be reformatted to a great extent, they are fairly responsive to changes in science community needs, he said.
The Integrated Ocean Drilling Program—a prior incarnation of the current IODP—had the same capabilities, but operated under a different structure. According to Clement, concerns of the three programs becoming independent and drifting apart resulted in the creation of a nonprofit central management organization (IODP Management International [IODP-MI]) charged with creating similar processes and procedures across all three platforms. Under that management system, he indicated, the IODP-MI gave a contract for a portion of operations to the Consortium for Ocean Leadership,1 which in turn awarded a subcontract to Texas A&M University, which houses the JOIDES Resolution, and to Lamont-Doherty Earth Observatory (LDEO) to provide oversight of wireline logging service subcontractors. He noted that this was a cumbersome model for U.S. facilities, and it proved difficult for IODP-MI to force a single pattern of operations onto all aspects of operations and onto organizations from multiple countries. Each of the three platforms had different technical capabilities, foci, and decision-making processes to implement science, he said. Money from multiple sources had to be tracked separately, making accounting more difficult.
At present, according to Clement, the JOIDES Resolution works under a cooperative agreement rather than a contract. Clement provided the following example as an advantage of this arrangement over that of the IODP-MI: a 2-day delay in refueling a vessel—a $2 million event—would have pushed the expenditure across a fiscal year boundary. Accommodating the event would have been difficult under a contract agreement, but the flexibility of the cooperative agreement made this a “non-event.” The former system was top-heavy and administratively difficult for those platforms at that time, he said. Operations were streamlined when funding became more constrained, resulting in the current drilling program management scheme. Most of the present funding for the JOIDES Resolution is from
___________________
NSF, but international partners also provide funding directly to NSF as unrestricted donations to JOIDES Resolution operations, making management of those funds simpler.
The JOIDES Resolution and other IODP platforms are overseen by a facility board that sets the schedule and priorities for operations. In the current program, the facility boards for each platform operate fairly autonomously, although all take advice from a similar set of advisory panels. Clement described the program as being a “bottom-up driven program written from the top-down,” indicating that the science community drives IODP. The JOIDES Resolution, he indicated, undergoes a multi-tiered assessment process including cruise evaluations and feedback to everyone who participated on a cruise, an annual 2-day co-chief review of expedition implementations, and an annual review by an NSF panel. The NSF panel includes members of the ocean science and related communities and directors of other large NSF facilities. Based on that review, NSF provides guidance to the JOIDES Resolution facility and board. The program reprioritizes its resources in response, as appropriate.
Clement described how scheduling priorities are set. The program is nominally funded to operate four or five expeditions per year. At the time of the workshop, he said that more than 100 unsolicited drilling proposals had been submitted, of which four will be funded. International panels of scientists from member countries review the proposals. They review for science quality and whether the proposed science can be achieved at a given site. Highly ranked proposals are forwarded to the appropriate facility board and evaluated by a safety panel of industry experts to evaluate the safety of the proposed operations. Each platform’s facility board will then prioritize scheduling and operations, decide on the ship track, and consider efficiencies, Clement continued. The platform then invites the shipboard scientists to participate in expeditions to assure quality data collection.
Clement explained that IODP operates under a 10-year science plan that outlines the priorities for the international program. Each proposal must be responsive to those priorities. Proposal review and feedback processes are more streamlined and nimble than in earlier years. Investigators provide a letter of intent and quickly receive feedback regarding whether a full proposal is warranted, he noted. The program has also improved its processes so that experiments generally are implemented within 2.5 years of being proposed.
Clement observed that the Sea Change: 2015-2025 Decadal Survey of Ocean Sciences2 report directed NSF’s Division of Ocean Sciences to better balance IODP facility and science costs. In response, there was an effort to separate facility and science operations, and NSF began funding the JOIDES Resolution through one NSF channel and associated science-related costs (e.g., support of U.S. scientists on an expedition and their post-cruise research) through another. This resulted in more efficient science reviews by panels.
Clement described that, overall, the current structure works well, and the three heavily involved universities benefit from such a large program. He noted the unintended consequences of splitting funds so narrowly, particularly in the United States where functions are distributed among multiple universities. (Texas A&M University has the contract for the JOIDES Resolution, LDEO runs the U.S. Science Support Program, and the Scripps Institution of Oceanography runs the Science Support Office.) For example, the JOIDES Resolution receives no funding for education and outreach, but demand can arise for use of the ship for these purposes (e.g., when arriving at a port of call and several hundred people want to tour the ship). He said that managers do try to facilitate such activities in spite of these barriers.
___________________
2 See https://www.nap.edu/catalog/21655/sea-change-2015-2025-decadal-survey-of-ocean-sciences.
MANAGEMENT OF THE NATIONAL ECOLOGICAL OBSERVATORY NETWORK (NEON)
Michael Kuhlman, Battelle
Michael Kuhlman is the chief scientist of Contract Research at Battelle. Battelle is a nonprofit 501(c)(3) organization with one part that manages six Department of Energy National Laboratories and one Department of Homeland Security National Laboratory, while another part is the contract research organization that performs approximately $800 million per year in contract research for government, commercial firms, and others. Kuhlman works for this latter part of Battelle, which competed for management of NSF’s NEON in 2015.
Battelle entered into a cooperative agreement with NSF in 2016 to manage NEON—an ecological observatory network with 81 sites in 20 eco-climate zones in multiple states and Puerto Rico. These produce 177 different terrestrial, aquatic, and aerial data products that support regional-, continental-, and global-scale ecological research, Kuhlman said. The network is intended to operate for a 30-year period. NEON provides data, protocols, education and training materials, and assets to the ecological research community. Examples provided by Kuhlman of NEON observations include exchange of carbon dioxide from the surface to the atmosphere (and vice versa), biomass buildup during the year, and aerial observations such as Light Detection and Ranging (LiDAR) and hyperspectral imaging. He noted that data, protocols, and processing algorithms are freely open to any investigators in the world through the NEON portal,3 as are training materials and software tools developed for working with NEON data. Researchers are also provided access to NEON infrastructure and enhanced sampling at sites on a cost reimbursement basis.
Kuhlman described Battelle’s matrix management structure for both its technical performance and support functions. A large portion of the program staff is fully funded by the program, but about 230 people each year are employed during the active field season. The matrix structure allows Battelle to draw from the additional 3,000 research and support staff available to any of the approximately 900 contracts Battelle performs in a typical year. Despite this broad matrix structure, instances occur when field staff with the right qualifications are difficult to find. Generally, however, Kuhlman said that the matrix structure allows for the ebb and flow of people and experts from project to project within Battelle, including human resources, contracts, financing, and purchasing. Charges to NEON program accounts are managed through Battelle’s cost account managers, who are responsible for delivery on schedule and budgeting of their portions of the scope, as agreed to in the NEON Annual Performance Plan. According to Kuhlman, the management of NEON has not changed significantly since Battelle took over, but some of the structure has been streamlined as construction has wound down.
Figure 4.1 is a NEON organizational construct presented by Kuhlman. A chief scientist for the program sets scientific priorities as reflected in their Annual Performance Plan (NEON is evaluated by NSF in an annual site review). The chief scientist is assisted by an operations manager who is responsible for the “how” and “when” of accomplishing priorities within budget and on schedule, he said. Six functional program elements make up the lower tier of operations (Science, Field Science, Field Support, Instrumentation, Data Infrastructure, and Engagement). The instrumentation and science teams are responsible for implementing new instrumentation and best practices. Kuhlman stated that providing data is NEON’s biggest form of user support, and that support comes largely through Battelle’s data infrastructure and engagement teams.
___________________
Kuhlman indicated that NEON has a strategic engagement plan that exceeds funding availability. The chief scientist prioritizes among activities in the plan. The engagement team has additional funding for activities (such as hosting a workshop aimed at increasing diversity in the data sciences in the environmental field). Permanent staff at each facility interact with site hosts and other researchers working at the sites. Staff also interact with the local education communities through lectures, training, seminars, data access, and assisting students with performing projects using NEON data. Battelle is in the early stages of implementing a tool to capture these types of activities in consistent ways.
Education and outreach are driven by the engagement team under the direct supervision of the chief scientist. Workforce development, focused on creation and formation of the next generation of ecologists, is largely an operation for the field science teams and scientific staff, along with the engagement teams. Figure 4.2, shared by Kuhlman, is a partial responsibility assignment matrix for facility operations. A program director, incorporated into the structure, resolves any disagreements regarding program implementation that may arise between the chief scientist and observatory director. The program director will identify a common course forward, then present that course to NSF for concurrence and to the community.
Kuhlman stated that the management of NEON is informed by that of other Battelle facilities. The manager of Battelle’s National Laboratory Operations reports directly to Battelle’s Chief Executive Officer, as does the NEON Contract Research Manager. Management of NEON and the National Laboratories have similarities, but additional requirements are imposed on the management of the National Laboratories. The directors of the National Laboratories each report to the respective boards responsible for the operations of each Laboratory. In all cases, there is an attempt to minimize the number of intermediaries between the functional organizations and the director.
NEON was conceived and initiated through a community-wide effort prior to Battelle’s commencement of management responsibilities. According to Kuhlman, Battelle assumed this role with NEON when just under 25 percent of the construction program was complete. Battelle completed the remaining construction, and is now entering into the initial operations phase for the entire observatory. Battelle implemented an earned value management approach4 and a bottom-up budget for each structure element, as well as an integrated master schedule to shift resources within and across the program as needed to maintain workflow. Battelle also implemented processes to manage change within the program that allows for flexibility and compromise so that the forward trajectory may be adjusted as appropriate. According to Kuhlman, NEON uses a Project Change Control Board that includes people from the parts of the program that either are impacted by or need to accommodate the change. Recommendations from the Project Change Control Board go to the Chief Scientist and Operations Manager.
According to Kuhlman, the cooperative agreement between Battelle and NSF calls for interaction with stakeholders through external advisory bodies. Battelle receives strategic advice regarding scientific and technical matter and direction through a Science Technology and Education Advisory Committee (STEAC). Kuhlman said that Battelle sends out a call for applications to serve on the STEAC. The applications are evaluated and voted on by the STEAC and accepted by Battelle. Approximately 12 STEAC members have a range of skills and serve 3-year terms. Battelle informs NSF of guidance received from the advisory committee and Battelle’s response and rationale for the response.
___________________
4 A technique for measuring project performance by combining measurements of project scope, schedule, and cost.
According to Kuhlman, additional tactical advice is received from 18 functional Technical Working Groups (TWGs) that are organized by topics spanned by the observatory. About 170 members were selected to serve in the TWGs, he indicated. They interact closely with Battelle science and domain staff and consider topics such as data protocols, best practices, and new instrumentation. Battelle receives further input on the science content of NEON through research community workshops, and through routine field staff interaction with local researchers, educators, and members of under-represented communities at field sites. Eighteen domain managers covering the 81 sites share information about effective practices, and materials are shared through a portal. He noted that NSF is developing plans for forming a user advisory group to NEON through the Directorate for Biological Sciences Advisory Committee.
The principal means by which NEON receives feedback, Kuhlman noted, is through the STEAC and through external community workshops. The STEAC is a body of external advisers to Battelle that meets monthly virtually, and twice per year in person, to develop and provide a strategic, longer-term vision and guidance for program evolution. They provide reports that Battelle shares with NSF, along with the actions taken by Battelle and responses about feasibility. The reports are posted online. STEAC’s principal recommendations to date have been related to engagement and outreach. Battelle expects to receive more recommendations regarding science direction as the observatory matures. If STEAC makes a recommendation and NSF concurs, then Battelle will determine how to follow the recommendation. STEAC does not have authority to commit or direct resources on behalf of the program. However, the science team, through interaction with community members, will sometimes propose mechanisms for changes in focus or science direction to the advisory committee.
According to Kuhlman, Battelle currently assesses success in achieving community goals by determining whether NEON data outputs are consistent with goals established by the initial design. Battelle recognizes and has discussed with NSF that the initial design is 10 years old and may not be in complete alignment with current needs. Battelle statisticians study the data to determine the appropriateness of sampling designs. However, determining whether the observatory addresses community needs is currently a measure of numbers of publications and publication impacts, he indicated. Battelle depends on input from the STEAC and external community workshops to determine if the initial design meets demand.
MANAGEMENT OF NASA’S EARTH OBSERVING SYSTEM DATA AND INFORMATION SYSTEM (EOSDIS)
Jeanne Behnke, NASA Goddard Space Flight Center
Jeanne Behnke is the Earth Science Data and Information System (ESDIS) Project deputy manager for operations at the NASA Goddard Space Flight Center (GSFC). Behnke’s presentation described the history and ongoing management of NASA’s Earth Observing System Data and Information System (EOSDIS). She began her presentation by describing the evolution of the model for projects at NASA. In the 1980s, she said, a principal investigator in Earth sciences would develop all parts of a project, from instrumentation to data collection to analysis. An investigator would be contacted directly for any information or data about the project. NASA made changes to this under the Mission to Planet Earth Initiative5 and initiated an open data policy to promote data usage by the community, including international partners.
___________________
EOSDIS was originally designed to be a multi-mission operation for the Mission to Planet Earth platforms—including the Tropical Rainfall Measuring Mission6 and the Terra,7 Aqua,8 and Aura9 missions—composed of many functions including spacecraft operations, downlink receipt of data to Earth, processing, archiving, and distribution. Every NASA Earth science spacecraft developed since these changes were implemented was to be part of the EOSDIS system and the associated archive and distribution system would also be used for other data. Eventually, according to Behnke, the spacecraft operations component was moved elsewhere. EOSDIS now includes only interaction with data providers and users in the form of data processing, archiving, parsing, and distribution of data from all NASA Earth science missions. Behnke noted that at present, EOSDIS is a multi-mission operation system including about 95 missions and 30 petabytes of data in 438 million individual files. It delivers 1.6 billion data products to more than 3.1 million users around the world. Data are now available to users within 3 hours of satellite acquisition. EOSDIS has interoperable data archives, does science data processing and management, data access online through a common tool, and data transport to handle large downloads, she said.
Behnke described the presence of 12 discipline-oriented distributed active archive centers (DAACs) across the United States, operated by about 600 employees. The DAACs are key components of the EOSDIS charged primarily with the archiving and distribution of NASA’s Earth science data collection. The metrics mentioned above are an aggregate of the work of all of the DAACs. She stated that some DAACs may need $10 million per year to operate while others may cost less than $1 million. These budgets depend on the volumes of data, traffic, data services, and sizes of individual data files. The ESDIS Project is able to shift funds and resources at the project level to cover, for example, staffing needs.
Behnke said much interaction occurs among the DAACS; for example, the DAACs may collaborate to bring systems engineers, user services groups, or outreach teams together for meetings, webinars, or online media activities. She stressed that communication across the entire science community can be difficult, so each DAAC has a user working group—a group of science users—that communicates throughout the year with its assigned DAAC and prepares a report to highlight data management, data collection, or science tools issues within the DAAC. This report is sent to the ESDIS Project and to NASA Headquarters for review and incorporation into the EOSDIS system.
According to Behnke, ESDIS conducts annual performance and budgeting exercises with NASA Headquarters and with GSFC. The EOSDIS is federally managed through contracts and interagency agreements, as well as working agreements with other NASA centers. The ESDIS Project provides user access to data from NASA’s Earth science missions through the development and operation of the sciences systems of EOSDIS. The design of EOSDIS is informed by many stakeholders, including but not limited to the NASA Headquarters Earth Science Division, science teams, the Mission Operations Centers (e.g., providing information about new geolocations that need to be accounted in data processing as satellites age and deorbit), and the NASA Chief Information Office.
Behnke highlighted the Oak Ridge National Laboratory DAAC (a partnership with the Department of Energy), where approximately 1,400 individual data sets related to bio-geochemical dynamics, ecological, and environmental processes from NASA-funded field campaign studies are archived and made available to all researchers. She also discussed the Alaska Satellite Facility (ASF) DAAC at the University of Alaska Fairbanks, where synthetic aperture radar products and data related to sea ice, polar processes, and geophysics are
___________________
6 See https://pmm.nasa.gov/trmm.
7 See https://terra.nasa.gov.
8 See https://aqua.nasa.gov.
housed. The ASF also mirrors data from the European Space Agency’s (ESA’s) SENTINEL10 program at that facility as part of a NASA/ESA bilateral agreement.
To demonstrate the volume of activity at a DAAC site, Behnke stated that the ASF DAAC distributed almost 65 terabytes of SENTINEL data in a recent 1-month period. During that month, more than 2,000 users from Brazil, Chile, Columbia, and Peru were registered, indicating use of the data for projects around the world. Users rely on the DAAC centers not only for data, but also for user support. Behnke stated that all indicators suggest that data sets will get larger in the near future. The current “on-premises” type of model employed by each DAAC will not be able to keep up with data archiving and demand, and more user support will likely be necessary, she noted. Given the volumes of data, NASA Headquarters recently directed EOSDIS to study and plan to move data operations to the commercial Cloud.
Moving to the commercial Cloud requires many new processes while continuing to operate under NASA’s open data policy. For example, ESDIS has had to create software to avoid potentially expensive data egress charges from Cloud providers. In-house training was developed for ESDIS resource analysts to understand how to monitor cost status, Behnke continued. NASA Headquarters is also brokering discounts for the Agency with the commercial Cloud providers based on the amount of storage they expect to need, which means that ESDIS needs to provide accurate data volume estimates for this and future years. Plans are in place to test commercial Cloud applications by migrating 30 of their most popular products, she said. Those data will be available via the Cloud but also remain on premises until the data in the Cloud are deemed stable. As data are migrated to the Cloud, the DAACs need to encourage science users to do their processing within the Cloud framework. This means encouraging the use of various machine-to-machine interaction techniques (e.g., scripts). NASA Headquarters is also pushing for more data system software to be open source. Behnke mentioned that building application programming interfaces is important to the user community.
The Earth Sciences Data Systems at NASA Headquarters provides funding and direction to the project. Behnke described a system architecture model and listed the following sources of advice and user feedback:
- Opportunities for EOSDIS to test new interfaces, technologies, or methods from specific programs;
- EOSDIS staff communication with the science teams of new NASA missions to better understand data ingest and archive requirements;
- The National Academies of Sciences, Engineering, and Medicine’s 2017-2027 decadal survey conducted for NASA (Thriving on Our Changing Planet: A Decadal Strategy for Earth Observation from Space [2018]);11
- Information technology and security directives from the Chief Information Office (which are becoming more difficult to accommodate as the program ages);
- Other agencies, such as the U.S. Geological Survey and the National Oceanic and Atmospheric Administration, and international partners;
- The ESDIS12 project that is managed within EOSDIS system architecture;
- The American Customer Satisfaction Index, an independently contracted survey; and
- Science researchers and applications users.
___________________
10 See https://sentinel.esa.int/web/sentinel/missions.
MODERATED DISCUSSION WITH FACILITY MANAGERS
Following the presentations, Behnke, Clement, Detrick, and Miller sat on a panel and took questions from the workshop participants. The workshop planning committee’s goal for this discussion was to provide a common understanding among workshop participants of the commonalities and differences among the management, governance, and decision-making models across the different facilities. The next sections summarize some of the key themes of the discussion.
Differences in Management Styles
Detrick stated that he was struck by the difference in management structures, governance approaches, and methodologies for community interactions amongst the different facilities. He summarized some of the differences by highlighting that IODP is an international consortium with different countries supporting different aspects of the program, and with evaluation of science proposals done internally, while the Incorporated Research Institutions for Seismology (IRIS) and UNAVCO are dependent on NSF to make science funding decisions. Detrick noted that each facility has adapted their operational models to suit their needs. He continued by stating that the matrix management model employed at NEON could work well with a large organization such as Battelle, but that IRIS and UNAVCO do not have sufficient staff resources to implement such a model. A workshop participant added that the size of the research community and its expectations also drive the management models.
Clement noted there are circumstances when it is practical to organize a facility as a corporation, but other circumstances for which the infrastructure of a university is beneficial. He cited being able to take advantage of the university’s general counsel and human resources programs as a source of potential cost savings, although he did note that university-imposed requirements can impede facility processes. Clement said that a cross-facility examination of relative costs might be informative.
Jon Alberts, executive secretary of the University-National Oceanographic Laboratory System (UNOLS), gave a brief overview of UNOLS as another example of a scientific facility management structure. His comments are summarized in Box 4.2.
Joint Facility Committees
Detrick clarified that the structure of the Joint IRIS/UNAVCO Executive Committee, described by both Detrick and Miller in their presentations (see Chapter 3), was still being planned and implemented. He said that IRIS and UNAVCO subcommittees might meet face to face to address cross-cutting issues, but might also meet separately to handle issues unique to each facility. Meertens added that IRIS and UNAVCO facility managers have always participated in each other’s committee meetings. Aster pointed out that joint IRIS/UNAVCO committees are not new; the IRIS Polar Network Science committee, for example, was established as a joint IRIS/UNAVCO committee. He said that the committee works well to address issues common to both facilities, and is effective at addressing issues of research communities without their own consortia (e.g., the glaciology community).
Effect of the EarthScope Plate Boundary Observatory on IRIS and UNAVCO Management
A workshop participant asked how completion of the Plate Boundary Observatory (PBO) and the EarthScope movable array impacts UNAVCO and IRIS. Miller responded that the biggest internal issues nearing the end of construction were related to staffing. Staffing levels had to be cut substantially when construction was over, but there was a need to incentivize staff to stay until the very end of construction. By realigning resources dedicated to field applications, geodetic infrastructure, and data, UNAVCO was able to restructure and move from a construction mode into an operations and management mode. With respect to future operations, Miller stated that UNAVCO is developing a strategy to integrate the PBO with networks constructed by UNAVCO in Mexico and the Caribbean under independent awards with NSF.
Detrick noted that the EarthScope movable array experiment continues, with subcontractors handling data processing. IRIS is in a similar stage that UNAVCO was in prior to the transition from construction to operations and maintenance. Instruments are to be removed completely by 2021, when the transportable experiment will end. No follow-on experiment is being planned. According to Detrick, IRIS will disband the group, perhaps incorporating some staff into other IRIS operations. It will have to restructure, much like UNAVCO did. IRIS successfully developed a new capability—the transportable array—but IRIS will likely lose the people implementing the array.
Facility Reviews and Feedback
According to Detrick, NSF’s National Science Board instituted a policy that facility awards were to be renewed or re-competed every 5 years, depending on the circumstances around the facility. The 5-year proposal renewal is, in itself, a review. Individual IRIS and UNAVCO facility components may be reviewed more frequently. Miller pointed out that the facilities also undergo a management review every 5 to 10 years. Clement added that large facilities that receive funding from NSF’s Major Research Equipment and Facilities Construction account undergo additional oversight to ensure that all aspects of the program are legal and appropriately documented. Meeting these requirements, noted Clement, could be challenging for a smaller facility such as UNOLS.
A workshop participant asked the panelists about approaches used to assess how well their respective boards provide guidance. Detrick responded that IRIS has no separate mechanism to directly assess performance of the board over time, but suggested that the periodic reviews of program elements, the contract renewal process, and NSF evaluations provide indirect information about how well the Board of Directors is doing as the body with principle responsibility for the organization. Clement responded that because operations of the JOIDES Resolution are embedded in a university, it does not have a board of directors. It does, however, have a facility board that includes representatives from each of its funding agencies, plus six scientists (three from the United States and three representing international partners). As part of the annual review, an NSF panel evaluates how effectively the facility board provides direction to the facility.
A participant noted that facilities that have boards of directors from the user community (e.g., IRIS and UNAVCO) have built-in mechanisms for successful incorporation of community feedback, but wondered how other facilities without such boards measure success. Some obvious metrics (e.g., numbers of publications) may take too long to measure to be useful. Another participant indicated that success might be measured by the number of new scientists coming into the field and serve on advisory committees, but this also requires a long time to measure. Detrick suggested the response to Board of Directors recommendations could be on a shorter timescale, again raising the example of the nodal sensors he described earlier (see Chapter 3), which occurred in only 3 years. If IRIS actions were not responsive to community needs, Detrick stated, IRIS standing committees would communicate with the IRIS Board of Directors, who, in turn, would develop a plan with facility management to address the given issue.
Implementing Recommendations and Managing Expectations
Clement described a project portfolio management process at the JOIDES Resolution facility that allows for assessment of benefits given a request to change instrumentation or data systems. Facility management considers who benefits from the change (e.g., internal or external customers), whether the change will affect one or multiple expeditions, or one or multiple scientific needs. Clement continued by describing that facility management then looks at available resources, and where changes could be implemented to the best satisfaction of the community. Detrick stated that the IRIS Board of Directors makes final decisions on budget priorities, and gives guidance to the standing committees. The standing committees, in turn, work with program managers to implement the priorities.
When asked how facilities manage scientific community expectations, Detrick responded that having IRIS governance committee members involved in strategic planning and budget prioritization is an effective way to manage community expectations. Those committee members make the difficult decisions, and then communicate the realities of what can and cannot be done to the broader community.
Training the Board of Directors
Miller explained that the UNAVCO Board of Directors created a training and development plan for board members that includes a rotation of topics over a 2-year cycle. The plan ensures that the board understands the full range of responsibilities and fiduciary oversight. UNAVCO takes advantage of external resources, and sometimes brings in external people to conduct the training. Detrick described that IRIS has three new board members join the board each year. At that time, IRIS has a 2- to 3-hour training session for new board members that includes discussion of both legal and fiduciary responsibilities of the board and individual board member responsibilities. On reflection, Kuhlman added later that training sessions for NEON board members has not been part of NEON operations, but that it would be something NEON would adopt, moving forward.
