Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 43
5 Pathways from Research to Operations Over the past several decades, many NASA research capabilities have been successfully transitioned to NOAA operational use, and NASA and NOAA have developed a variety of processes to accomplish these transitions (Obermann and Williamson, 2002). However, there is no overarching or consistent set of processes covering all transition activities. This chapter summarizes lessons learned from 10 case studies, and then describes NASA-NOAA transition approaches and compares these with approaches used by other U.S. agencies and foreign space organizations. Transition pathways between NASA and NOAA will continue to require a wide-ranging set of approaches, as each research concept and operational need will be associated with its own matrix of both opportunities and obstacles. NASA and NOAA have worked for more than two decades to improve these modes of transferring research into operations,1 but have been faced with “too many degrees of freedom” in trying to match the NASA research and NOAA operational programs as a whole.2 In the past 10 years, NASA has focused on climate monitoring and, consequently, separated itself to some extent from its historical role of transitioning 1 For example, the 2001 NESDIS (National Environmental Satellite, Data, and Information Service) Strategic Plan includes the following Objective 1.6: “Take advantage of opportunities for transition of remote-sensing technology developed by NASA for research that meets NOAA’s operational needs” (NOAA, 2001d). 2 Presentation by Ghassem Asrar, Associate Administrator, NASA Earth Science Enterprise, to the Committee on NASA-NOAA Transition from Research to Operations, January 2002, Washington, D.C.
OCR for page 44
weather prediction research capabilities to NOAA operations.3 While the relationship between NASA and NOAA is “excellent,”4 both agencies recognize that current processes for transitioning research capabilities to operations could be improved. CASE STUDIES AND LESSONS LEARNED The committee examined 10 case studies of transitions, the details of which appear in Appendix B, “Case Studies of Transitions from Research to Operations.” These case studies were chosen because they contained lessons learned in transitioning research to operations. A summary of the lessons learned is presented in Table 5.1. (Obermann and Williamson  discuss other case studies and similar lessons learned.) A common theme of these studies was the need for a management structure and a formal set of processes that could speed the transition of research to operational use. In the case of the early days of the Defense Meteorological Satellite Program (DMSP), an efficient, compact management structure and process allowed the program to far exceed expectations and outpace parallel efforts in NASA. In the case of the infrared sounder, NOAA did not fully exploit this instrument’s potential for more than 25 years after the sounder first flew, because of resistance to change and lack of knowledge about how to use the observations effectively in numerical modeling. In the cases of the Volcanic Ash Mapper (VOLCAM), lightning detection, and the Solar X-ray Imager (SXI), no plan was in place to make the transition to operations until very late in the process, if at all. In the cases of the Special Sensor Microwave/ Imager (SSM/I), Very High Resolution Radiometer (VHRR)/Advanced VHRR, and Tropical Rainforest Measuring Mission (TRMM), the involvement of the research and operations community early in the process was extremely important to their success. The value of research advocacy and operational involvement was evident in the ocean altimetry and scatterometry cases. Throughout all of these cases, the need for oversight at the highest levels of the transition process cannot be overemphasized. From these case studies, some general conclusions about transition pathways and their associated processes can be drawn. These are discussed in the following sections. 3 Presentation by Ghassem Asrar, Associate Administrator, NASA Earth Science Enterprise, to the Committee on NASA-NOAA Transition from Research to Operations, January 2002, Washington, D.C. 4 “NASA-NOAA Transitions from Research to Operations,” presentation by G.W. Withee, Assistant Administrator for Satellite and Information Services, to the Committee on NASA-NOAA Transition from Research to Operations, January 2002, Washington, D.C.
OCR for page 45
TABLE 5.1 Summary of Lessons Learned from 10 Case Studies Case Study Lessons Learned Infrared sounder The extended period between the first flight of a sensor such as the infrared sounder and the full operational use in numerical models can be attributed to many factors: resistance to change in the operational organization (e.g., insistence that the data look like radiosonde data); lack of a technology-transition plan or process (the operational centers were not prepared to use the data when they were first available); and incomplete research (the development of mathematical methods to use raw radiance values in numerical modeling took more than 25 years). Very High Resolution Radiometer (VHRR)/ Advanced VHRR (AVHRR) The operational agency (in this case NOAA) cared enough about the potential utility of the instruments to help fund their development. It is important to involve the research and operational community early in the process. The researchers communicated with potential users, educated them about the observations and their potential use, and “marketed” the mission. Near-real-time availability of research data to users is a valuable part of the transitioning process. Defense Meteorological Satellite Program (DMSP) The novel management scheme was made possible by a small program office, consisting of a few key energetic people with strong ties to the user (USAF Air Weather Service) and the research community (University of Wisconsin, Madison, primarily), that exercised technical direction. It could make decisions and act quickly. The office achieved an excellent success record at low cost. Incorporating the needs of the user into early instrument design and data-processing systems pays big dividends in transitioning systems from research to operational use. Involvement of the academic community in missions can lead to stronger scientific underpinnings, help infuse science with new ideas and concepts, and educate and train the next generation of space scientists and engineers. Lightning detection from space No transition pathway was established, in part because there was insufficient push from the research community and pull from the operational community. As a technology advances past a proof-of-concept stage, there should be a parallel investment made to determine its viability for use by either direct or intermediate users, to push the development of an operational requirement for the proven technology. The result of the proof-of-concept demonstration must be a documented user requirement for the new observation capability in order to create the pull needed to transition the experimental capability into an operational measurement system. Ocean altimetry There can be a positive synergy between research and operations. Apparently conflicting science and operational requirements can be overcome through communication and leadership at all levels, including administration and the research community. For example, the research community continued to show the value of ocean altimetry and how continuous improvement can lead to new insights. The highly visible success of TOPEX/ Poseidon encouraged senior management to take risks and seek new initiatives for follow-
OCR for page 46
Case Study Lessons Learned on missions. The operational community recognized the value of these measurements and was willing to examine new technologies such as wide-swath altimetry to continue to expand the understanding of ocean circulations. Scatterometer Near-real-time data return from research missions is required in order to interest the operational numerical weather prediction agencies. Although graphical products from research missions can be produced rapidly for subjective forecasting use, the assimilation of new observations into numerical weather prediction forecast/analysis systems requires detailed knowledge of the measurement error characteristics and testing in the operational setting. This has generally not been achieved for at least 1 or 2 years after calibrated geophysical data become available from new instruments. Test beds such as the NASA-NOAA Joint Center for Satellite Data Assimilation can play a significant, positive role in speeding the transition of research to operations. Notwithstanding successful exploitation of scatterometer data by meteorological researchers and operational use of scatterometer measurements (acquired by both U.S. and international missions), the incorporation of surface vector wind measurements into the operational observing constellations has taken 10 to 15 years and is not yet a reality. Special Sensor Microwave/ Imager (SSM/I) It is important to consider how data will be used well in advance of a launch, and to develop algorithms that will process the raw data to generate useful products. The responsibility and funding for delivering sensor data in a form useful for operational users should be borne by the space segment of the process. The space segment cannot be allowed to have the limited responsibility of only acquiring hardware; it must also have an integrated responsibility that includes continuing calibration and validation of SDRs and a baseline of EDRs. The full release and open availability to the research community of new operational data in both SDR and EDR form provide significant benefit through development of improved and new uses of the original data and feedback to the operational users. Solar X-ray Imager (SXI) Inadequate financial and human resources can prolong the transition from research to operations for many years after a technology has been demonstrated and a need established. Funding difficulties for SXI prolonged the transition process to nearly 30 years. Volcanic Ash Mapper (VOLCAM) VOLCAM was developed out of a long-standing interagency collaboration at the working level, a result of the NASA-led geophysical science and natural hazards programs’ having produced information of value to the operational agencies. Nevertheless, collaboration between agencies at the working level was not sufficient to establish agency commitment for the VOLCAM transition. NOAA has a very limited capacity or budget to evaluate new sensing concepts internally, so advancements in observational measurements are difficult to make unless they involve extending the capabilities of NOAA’s few core instruments.
OCR for page 47
Case Study Lessons Learned The lack of an organizational transition mechanism between NASA and NOAA makes direct transfer of technology between the agencies difficult. Other than the GIFTS mission, NASA has not funded any geostationary sensor proposals in recent years. Opportunities to evaluate new research sensors or measurements for geostationary operational use have thus been extremely limited. Tropical Rainfall Measuring Mission (TRMM) Availability of research data in real time fosters use and testing by operational users. Research data can have a positive impact on operations. Feedback from operations has a positive impact on research. NOTE: Details of the 10 case studies are presented in Appendix B, “Case Studies of Transitions from Research to Operations.” TRANSITION PATHWAYS AND PROCESSES As noted, NASA and NOAA share the motivation and need to improve the research-to-operations transition process. The need encompasses new products, including (1) technologies, (2) instruments, (3) measurement techniques, and (4) data products and data systems. An end-to-end set of processes for achieving transitions, whether formal or informal, is a transition pathway. Transition pathways vary depending on the science, technology, and applications involved, but they all contain building blocks, which constitute the infrastructure of the pathway. Building blocks may include, for example, a solid research foundation, laboratories, equipment, computers, algorithms, models, information technologies, and test beds. A transition pathway might begin with basic research supported by NSF or NASA. Based on sound theoretical and engineering principles, the pathway could include demonstration or proof-of-concept experiments that lead to the design of a NASA research mission. Ideally, research missions would include an overarching architecture, or plan, for the transfer of the research to operations. The transition pathway would include research on the necessary methods of using the data in numerical models (i.e., data assimilation) as well as in other uses, such as monitoring climate. Toward the end of the transition pathway, test beds would demonstrate the use of the research data in near real time, evaluate their operational impact, and provide feedback to the researchers. After a demonstration of positive impact, future operational missions would produce the data in real time, and operational centers and other users would incorporate the new data into their operations, completing the transition pathway.
OCR for page 48
Transition pathways can be characterized in terms of four elements: Objectives—What is the purpose of the transition process, and what are its objectives? Is there a plan for a complete end-to-end set of processes (a transition pathway) that will achieve the objectives? Organizational structure—Who or what has the authority and responsibility for achieving the objectives? Is the authority effective, and is it adequately supported by the sponsoring agencies? “Organizational structure” includes the mechanics of formal reporting as well as the definition of the functional roles and responsibilities at every level of the organization. Necessarily, this implies a degree of “corporate” shared understanding; an individual’s understanding of his or her own job description alone is insufficient. Rather, each individual must understand how his or her roles and responsibilities relate to the mission and goals of the organization. Procedures—What processes exist for establishing the necessary plans, schedules, and resources? What are the plans, and are they sufficient? Are requirements carefully developed and adequately documented? The set of organizational procedures or way of operating is the “machinery” that must be run, maintained, and occasionally redesigned to enable an organization to produce research results for transition or to create products for particular applications. The organizational procedures include guidance, documents, rules, processes, communications systems, measures of effectiveness, and tools for assessment that will, if designed correctly, optimize the transition pathway. The challenge is to design these procedures consistent with the requirements of a bureaucracy (e.g., rules regarding personnel or acquisitions), while remaining flexible enough to accommodate agency mission requirements. Resources—Are adequate resources available, including personnel, funding, and schedule? Because they are subject to the executive and legislative budget processes, resources that may be inadequate often determine the limits on research and operational opportunities. These resources include the dollars available for research and operations (including any limitations on the reprogramming of such distributions) and the availability and distribution of appropriately educated and trained people. Additionally, an agency’s infrastructure (adequate laboratory space, necessary scientific and technical equipment, computational hardware and software) represents another important resource that factors in to both opportunities and obstacles to transition. TRANSITION CHALLENGES As with any complex process, transition pathways from research to operations and applications are characterized by a variety of challenges and potential barriers. No single barrier is the primary limiting factor in the ability to transition research to
OCR for page 49
operations, and no simple cure or “silver bullet” can eliminate all of the various impediments to successful technology transfer. Rather, these factors often combine with varying degrees of significance for any particular case to produce a formidable barrier to successful transition. Table 5.1. and Appendix B provide examples of transition needs that for various reasons have not been met. NASA selected VOLCAM as a flight candidate under the ESSP program in part because of its own expectation that the research flight of this instrument would lead to a demonstrated operational capability. But NOAA was unable to provide the long-term commitment required by NASA prior to approving the research mission, and so the research capability has not been successfully transitioned into operations. The Lightning Imaging Sensor (LIS) has flown on TRMM and appears to have support for transition, but it is not currently planned for operational use. Spaceborne tropospheric wind measurements, one of the highest-priority requirements of the operational community, have achieved neither research nor operational status owing to disagreement within the community about what research is needed, how it should be transitioned, and whether or not wind measurements should be purchased through the private sector. The limiting factor in transitioning research to operations can be inadequate scientific understanding or the difficulty of extending scientific understanding and/or technological capability to operational utility. There may be limits to the observing technologies, to the understanding of how to use the observations effectively (as in the case study of the infrared sounder), or to the computational power required to use the observations in operational models. A critical element of operations is the ability of a system to perform routinely under adverse environmental conditions—that is, to have operational robustness. Until new capabilities develop in other research sectors (e.g., computer science or mathematical analysis), a particular technology might be shelved rather than transitioned. This translates into a need to ensure that those programs engaged in advancing related technologies (e.g., materials or information technology) are kept aware of the impacts of their developments on the feasibility of transitioning environmental remote sensors from research into operations. Traditional processes and expectations often form impediments to instituting changes of the magnitude required to improve NASA-NOAA transitions. NASA has historically focused on the internal development and implementation of large projects. NOAA has tended to set long-term goals and requirements for its observing system capabilities, with limited flexibility for adapting to changing constraints or to the emergence of new capabilities. It appears that both organizations recognize these issues and are working to improve their flexibility. The inability to deal with massive volumes of remotely sensed data streams may be a technological barrier to transitioning. On July 24, 2002, the House Science
OCR for page 50
Subcommittee on Environment, Technology, and Standards recognized the risk associated with the $6.5 billion NPOESS program, scheduled to begin operating in about 2008. The subcommittee heard from a General Accounting Office (GAO) report that NOAA lacked a robust plan to handle the approximately 200 terabytes of data per year that NPOESS will deliver. Congressmen Vernon Ehlers (R-Mich.) and Mark Udall (D-Colo.) hit the nail on the head: Given these huge investments and the importance of satellites to so many aspects of our lives, it is our duty to ensure that the taxpayers are getting their money’s worth. But getting our money’s worth is not simply contingent on a satellite being successfully launched and data being beamed down. The key factor is being able to use the data. (Ehlers, 2002) NOAA’s investments in satellite technology have vastly expanded our ability to measure features of the earth, the oceans, and the atmosphere. Parallel progress in data processing and management is enabling us to utilize these measurements to improve weather forecasting and to better understand the global environment. The good news is we have a lot of data. On the other hand, the bad news is we have a lot of data. We must ensure that investments in technology and research to manage information keep pace with our investments in technology to gather it. (Udall, 2002) Thus, there is need for robust strategies to address the volumes, quality, and diversity of remotely sensed environmental data. There is a tendency to emphasize stand-alone observations rather than trying to optimize ways in which one set of observations might complement others. And, in some cases, there is a potential contradiction between the operational community’s eagerness for a robust data stream and the scientific community’s reluctance to compromise the accuracy or completeness of the data. Foreign collaborations are an increasingly important element of both research missions (e.g., TRMM, GRACE, Jason, CALIPSO, CloudSAT, COSMIC) and operational programs (e.g., NPOESS, Ocean Surface Topography Mission). (See also Appendix C regarding future missions.) These international collaborations—while valuable in many respects, including leveraging of scarce resources and sharing of scientific and technical expertise—often introduce complications such as national restrictions on trade and technology assistance, management complexities, and cultural differences. These complications can increase the difficulty in carrying out effective transitions. TRANSITION TYPES Within the current NASA-NOAA system, five primary research-to-operations types can be identified: (1) meteorological system upgrades, (2) algorithm and data product improvements, (3) NASA systematic measurements that transition to NOAA
OCR for page 51
operational measurements, (4) NASA exploratory measurements that transition to NOAA operational measurements, and (5) technology demonstrations that transition to operational systems. As described in the following subsections, each of these transition types is associated with a typical transition pathway, which has characteristic building blocks and processes. Meteorological System Upgrades Prior to 1981, NASA and NOAA cooperated effectively in developing new operational satellite systems. At that time, NASA typically funded “first unit” builds of weather satellites and their instruments and then transitioned proven capabilities to NOAA (and its predecessor, the Environmental Science Services Administration) for operational use. This cooperation was guided by a formal agreement established in 1973, the Operational Satellite Improvement Program (OSIP), which was funded at about $15 million per year. The budgets for NASA and NOAA reflected this agreement. NASA used its funding to develop prototype sensors, fly them on high-altitude aircraft, and transition them to research spacecraft for evaluation. Successful instruments were then provided to NOAA for transition to operational status. The program fell victim to NASA budget pressures and an Office of Management and Budget (OMB) desire to offload “routine” functions from NASA, and was canceled in 1982 (OTA, 1993; NRC, 2000a). Since 1982, there has been a formalized upgrade process for supporting NOAA sensor capability, primarily for the polar and geostationary satellite systems. Its objective is to provide incremental upgrades to existing sensors, major design upgrades to those sensors, and development and insertion of new sensors as needed. NASA has the authority, delegated from NOAA, to implement the procurement of the POES and GOES systems on the basis of NOAA requirements and budgets. Established processes exist for important parts of this activity, but they are largely derived from historical precedent developed over the multidecade POES and GOES collaboration rather than being carefully developed and documented procedures. Limitations in the processes, combined with budget or schedule constraints, have led to plans that have in some cases proven difficult to implement or have not been established in a timely manner (OTA, 1993; GAO, 1997). The POES program is currently being combined with DMSP into the NPOESS program,5 with NOAA and DOD as joint funding partners and NASA as a third unfunded partner focused on technology infusion. 5 ”NPOESS Risk Reduction: The NPOESS Preparatory Project,” presentation by Stanley Schneider, Associate Director for Technology Transition, NPOESS Intregrated Program Office, to the Committee on NASA-NOAA Transition from Research to Operations, April 2002, Washington, D.C.
OCR for page 52
The upgrade process has worked well for incremental upgrades (such as the evolution of the GOES imager channel selection), but it has been less successful with major advances or block upgrades that require the introduction of new systems or significant changes to sensors, spacecraft, or other mission systems (such as the introduction of the next-generation GOES sounder). Block upgrades have historically required technical, programmatic, and contractual decisions that limit the ongoing infusion of new capabilities. Technical decisions, such as the choice of spacecraft architecture for GOES-NEXT in the 1980s, can set constraints for many years on the nature of the new observational capabilities that can be accommodated. Programmatic or contractual decisions, such as the decision to procure follow-on units through fixed-price contracts on NPOESS, may minimize life-cycle costs, but they also reduce the flexibility to incorporate research- and technologically-driven advances as they become available. The IR sounder and scatterometer case studies (see Appendix B) are also examples in which the first type of transition, the system upgrade, did not work well. Time periods of up to 25 years elapsed before effective use was made of these technologies. In contrast, the DMSP and VHRR/AVHRR case studies show that this type of transition can work well. The key is effective and clear coordination between the operational and research entities. In summary, the objective in this first type of transition—meteorological system upgrades—is clear, and the structure and procedures are well documented. Overcoming the complications created by different agency cultures and obtaining the necessary resources are still a challenge. Algorithm and Data Product Improvements Operational employment of satellite data implies the use of the data by either an “end” user (usually a nonscientist) or an intermediate user (usually a scientist or a numerical model). An essential part of the transition pathway is the development of appropriate algorithms to transfer the raw measurements, or sensor data records (SDRs), into environmental data records (EDRs)—forms of environmental data that the user requires. Alternatively, algorithms can be developed that use the SDRs directly, as in data-assimilation schemes. The focus of the operational community has historically been on the EDRs. The direct use of the sensor records by the operational numerical models has only been a recent (last decade) development, after more than 30 years of satellite data availability. Observed cloud and precipitation data from any source (e.g., satellites or radars) have only recently been considered for input into numerical models. Two significant changes are occurring that have a major effect on the use of satellite data for operational purposes: first, the development of data-assimilation
OCR for page 53
systems that incorporate sensor data directly in Numerical Weather Prediction (NWP) models and, second, the acceptance of cloud and precipitation data as input parameters into these models. These changes shift the emphasis on operational use from focusing on EDRs to including SDRs as well. The significance of this evolution is that, whether in a “push” or a “pull” situation, the operational use of any proposed sensor data needs to be demonstrated. This demonstration should include the development of the appropriate algorithms, data-assimilation schemes, or other data products, to the point that soon after launch, the data can be used—in days to weeks, not months to years—by either the direct or the intermediate users. Therefore, a critical element in the transition of research capabilities to operational use is the development not only of the retrieval schemes to produce EDRs, but also of the assimilation methods for the proposed sensor data. In either case, parallel development and funding are required in order to achieve operational use of the satellite data. In the past, a lack of funding for the development of the necessary retrieval and assimilation algorithms has been a key reason for failures to quickly transition research satellite sensor data into operationally useful data. NASA and NOAA both perform research and development in the area of algorithms for operational applications and product generation. Recently, it was recognized that the complementary capabilities of the two organizations, the challenges to assimilate upcoming, large volumes of data, and the limited resources available to each agency provided motivation for establishing a joint capability. In response to these needs, the Joint Center for Satellite Data Assimilation (JCSDA) was recently chartered. With authority provided through an agreement between NASA and NOAA,6 JCSDA leadership has established clear objectives for the center as a facilitator in transitioning research algorithms and data products to operational status, has developed processes and plans for accomplishing these transitions, and has identified existing resources to support transition activities. JCSDA projects have included the transition of Quick Scatterometer (QuikSCAT) data to operational use. JCSDA is planning for similar transition of AIRS data and has selected GIFTS as a trial project. In summary, the objectives of this second transition type—algorithm and data product improvements—are clearly stated. There is an organization in place, procedures are drafted, and resources are becoming available. 6 “Joint Center for Satellite Data Assimilation,” presentation by Richard B. Rood, Director (Acting), Joint Center for Satellite Data Assimilation, to the Committee on NASA-NOAA Transition from Research to Operations, April 2002, Washington, D.C.
OCR for page 55
NPOESS system (including measurements not expected to be made within the current system procurement [the “unsatisfied EDRs”]) provides a sense of which exploratory measurements could lead to operational counterparts. However, there is no plan for linking exploratory missions to subsequent operational ones. The inability of VOLCAM,8 proposed as an ESSP exploratory mission, to develop a transition path is an example of the difficulty of this transition type. The SXI9 succeeded in making the transition, but only after considerable effort and strong commitment by individuals. The use of the peer-review process in selecting exploratory missions can be a transition impediment. While peer-reviewed selection is certainly appropriate for ESSP and similar NASA missions, the uncertainty associated with determining which measurement proposals will be selected makes planning by NOAA difficult. In summary, this fourth transition type—NASA exploratory measurements to NOAA operational measurements—is limited by a lack of clear, agreed-upon objectives and adequate procedures to define an efficient transition pathway. Technology Demonstrations to Operational Systems Although NASA’s role as a technology development agency for operational missions is clearly identified in programs such as NPOESS,10 NASA does not have a formal program for developing or demonstrating technology for NOAA sensors. Nevertheless, technology transitions are being performed on a case-by-case basis. The GIFTS instrument is being developed for flight validation under the New Millennium Program (NMP), with joint support from NASA and NOAA, in recognition of its value to the GOES-R sounder program.11 GIFTS was selected by NMP in 1999, with specific direction in the selection letter from NASA to establish closer ties with operational agencies. The mission for the GIFTS instrument has since been renamed GIFTS-IOMI, the operational content has been increased, and the mission now includes substantial contributions from NOAA and DOD. This transition type has evolved considerably as NASA and NOAA have worked to define the role of GIFTS in preparing for the GOES-R sounder. In particular, NOAA has targeted its contributions to provide for algorithm and data system modifications that support the assimilation of GIFTS data into operational systems of the 8 See the case study on the Volcanic Ash Mapper in Appendix B. 9 See the case study on the Solar X-ray Imager in Appendix B. 10 Presidential Decision Directive/National Science and Technology Council-2 (1994). 11 “Transitioning from Research to Operations: The GIFTS-IOMI Case Study,” presentation by F.Wallace Harrison, GIFTS-IOMI Project Manager, to the Committee on NASA-NOAA Transition from Research to Operations, April 2002, Washington, D.C.
OCR for page 56
National Centers for Environmental Prediction. It is less clear which aspects of the GIFTS sensor architecture or technologies will transition to the GOES-R sounder. In summary, the ad hoc evolution of the GIFTS-IOMI mission reflects the lack of an established and documented pathway for the fifth transition type—technology transitions from NASA to NOAA. The result is an unwieldy GIFTS-IOMI organizational structure with multiple partners having inadequately defined roles, procedures that have to be established as issues arise, and budget uncertainties that require each participating agency to plan resources without firm commitments from the other agencies. COMPARATIVE APPROACHES Research-to-operations processes in DOD, other U.S. institutions, and foreign organizations, as described in the following subsections, provide opportunities to compare transition approaches to those used by NASA and NOAA. Department of Defense The Department of Defense operates under a rigorous transition process based on operational requirements. The process is structured in a manner intended to optimize the formal definition of needs while maintaining a legally mandated arms-length relationship between acquisition and operations. It is also important to note that all of the research conducted by DOD is justified by and focused on the provisions of United States Code, Title X, consistent with the mission of the armed services. Within the individual services, the acquisition activities (including basic research, exploratory development, and advanced technology development— collectively defined as science and technology) are conducted under the secretariats (e.g., the Secretary of the Navy). The operations, which include the “demonstration and evaluation” components of research, are administered by the service chiefs (e.g., the Chief of Naval Operations). The process begins with the definition of mission needs and requirements by the operational forces. These needs and requirements are generally defined in broad terms (such as the need for an ability to assure access to any coastal area), but may include more specifics (such as the need for a near-real-time capability for characterizing the environment of that area) in support of operations. The Mission Needs Statement (MNS) can be an overarching document. More specific definition of requirements is developed through an Operational Requirements Document (ORD). In the joint arena in which multiple services are involved—analogous, say, to a NOAA and NASA relationship—a Capstone Requirements Document, or CRD, is the relevant document. The ORD is the working document that provides a very
OCR for page 57
detailed definition of what systems and platforms are needed and what their performance specifications should be. The MNS and the ORD are each subjected to formal review by the operational and acquisition leadership of the services. The ORD, once formally approved, is a “sacred” reference for the subsequent activities of planning, programming, and budgeting. It is this set of prioritized missions and requirements that is presented to the acquisition (including R&D) community. The system recognizes that the definition of an operational need might sometimes arise from emerging and unexpected technologies. The potential operational utility of radar, for example, came from research in physics, not from a specific requirement from the fighting forces. Thus, there must be some allowance for taking research risks in areas that are indirectly related to immediate operational requirements. Notwithstanding this robust process for connecting basic research with operational requirements, there is a potential downside. In the absence of strong tolerance for risk taking within the basic research, much of the fundamental science and technology can become too closely aligned with the immediate operational needs. In that case, the longer-term research investments could become marginalized, resulting in lost opportunities for introducing novel concepts that might translate to new operational capabilities. Within a structure such as that used in DOD, it is critical that there be a clearly stated policy of risk tolerance in research, as long as there is some potential payoff operationally. The challenge for the DOD research community is to assess the right balances of risk, relevance, and responsiveness in its research programs. Overall, however, the “system” expects that the research program will be in direct alignment with the operational requirements. The DOD approach provides all of the elements needed in a research-to-operations transition pathway—objectives, organizational structure, procedures, and resources. It is being used today in the NPOESS program. European Space Agency Transitions to EUMETSAT The European Organization for the Exploitation of Meteorological Satellites (EUMETSAT) is the European equivalent of NOAA/NESDIS. It is funded by 18 European member states to operate the European polar and geostationary satellite systems and to provide the satellite data to the weather and environmental services of the member states. The Global Monitoring for Environment and Security (GMES) initiative is a European Commission activity to coordinate geospatial information to support the decision-making process at European parliament and council of minister levels in environmental and security issues (ESA, 2000). Research and operational development is supported by the European Space Agency (ESA) for EUMETSAT. ESA provides funding for the first of a new generation
OCR for page 58
of satellites to meet EUMETSAT requirements, including Meteosat Second Generation and the Meteorological Operational Polar satellite (METOP 1). This arrangement is similar to that of the OSIP program in the early years of the NASA-NOAA collaboration. The objectives of ESA and EUMETSAT are clear, appropriate organizations are in place, the procedures are developed for the cooperative effort, and the resources have been acquired for these current programs. European Centre for Medium-Range Weather Forecasts The European Centre for Medium-Range Weather Forecasts (ECMWF) was established in 1973 to provide medium-range (out to 10 days) forecasts to the European member states. ECMWF has developed a well-defined pathway for transitioning research capabilities and data sources to operational status.12 The center uses a trial-model approach in which new data sources or capabilities can be evaluated through comparison with the same model without the trial data or algorithms. ECMWF makes less of a distinction between research and operations than is done in the United States, and is willing to incorporate research-quality data into operational systems before a thorough operational need or performance enhancement has been demonstrated. The center is also willing to incorporate research data into operations for periods of several years without assurance that the measurements will be continued. ECMWF has already successfully assimilated data from research sensors from the scatterometers on the European Remote Sensing satellites (ERS-1 and ERS-2) and TRMM and is planning transitions for the ESA Environmental Satellite ozone/ wind-wave data, AIRS, CloudSAT, and Jason-1. The culture of ECMWF has been to push the envelope. The organization is flexible and open to challenges, the procedures are in place, and resources are available for quick transitions. Office of the Federal Coordinator for Meteorological Services and Supporting Research The Office of the Federal Coordinator for Meteorological Services and Supporting Research, more briefly known as the Office of the Federal Coordinator for Meteorology (OFCM), is an interdepartmental office. It was established because 12 “ECMWF Responses to Inquiries from the NAS Study on NOAA/NESDIS Transition from Research to Operations,” presentation by David Burridge and Anthony Hollingsworth, European Centre for Medium-Range Weather Forecasts, to the Committee on NASA-NOAA Transition from Research to Operations, April 2002, Washington, D.C.
OCR for page 59
Congress and the Executive Office of the President recognized the importance of full coordination of federal meteorological activities. The Department of Commerce formed the OFCM in 1964 in response to Public Law 87-843—the Appropriations Act for State, Justice, Commerce, and Related Agencies (1963). The mission of this office is to ensure the effective use of federal meteorological resources by leading the systematic coordination of operational weather requirements and services, and supporting research, among the federal agencies. Fifteen federal departments and agencies are currently engaged in meteorological activities and participate in the OFCM’s activities. The OFCM carries out its tasks through an interagency staff working with representatives from the federal agencies. This infrastructure supports all of the federal agencies that are engaged in meteorological activities or that have a need for meteorological services. In addition to providing this coordinating infrastructure, the OFCM prepares operations plans, conducts studies, and responds to special inquiries and investigations. The successful development and deployment of the Weather Surveillance Radar-1988 Doppler (WSR88-D) and the Automated Surface Observing System are examples of multidepartmental programs successfully coordinated through the OFCM using the program council approach. OFCM program councils include high-level representatives from the appropriate participating agencies (for a particular project or process), coordinated and administratively staffed through the OFCM. However, the council members respond and provide budget commitments according to their respective agency procedures and structures. Integrated Program Office (IPO) Presidential Decision Directive NSTC-2 (1994) created the Integrated Program Office (IPO) and its oversight Executive Committee (EXCOM) to converge the NOAA and DOD polar-orbiting weather satellite programs into a national polar-orbiting operational satellite system, with the objective of reducing cost. The directive states that additional savings may be achieved by incorporating appropriate aspects of NASA’s Earth Observing System. The IPO is responsible for the management, planning, development, fabrication, and operation of the converged polar-orbiting satellite system. The EXCOM (formed at the undersecretary level) ensures that (1) both civil and national security objectives are satisfied; (2) program plans, budgets, and policies are coordinated; (3) agency funding commitments are equitable and sustained; and (4) the Senior User Advisory Group (with EXCOM representation) provides regular and frequent oversight of the IPO. NOAA has the lead for operations, DOD for procurement, and NASA for facilitating the development and insertion of cost-effective technologies.
OCR for page 60
In implementing its mission, the IPO generates Integrated Operational Requirements Documents (IORDs), which define the EDRs needed to meet the civil and national security missions. The IORD is signed by the EXCOM and forms the basis for all planning, program, and budget activity. Although the objectives for transitions are defined in Presidential Decision Directive NSTC-2 and the organization is functioning effectively, procedures for the transition process are not clearly spelled out, and the resources for transitioning have been pieced together on an ad hoc basis. TRENDS AND CHANGES IN TRANSITION PROCESSES NASA and NOAA are taking a variety of actions to improve the current transition pathways. NOAA’s National Environmental Satellite, Data, and Information Service has been reevaluating the architectures and requirements process for the entire meteorological observing system.13 This effort is driven in part by the near-term need for requirements that support procurement of the GOES-R series. NASA is undergoing a planning process, initiated in part by the mission plan established at the August 1998 Easton, Maryland, workshop held by NASA’s Earth Science Enterprise to assess candidate mission programs for the 2002-2010 period (Kennel et al., 1998) and furthered through ongoing dialogue with OMB.14 In response to the requirements for GOES-R, a Research Strategy and Strategic Plan (NASA, 2000) has been established by NASA, and mission road maps are being developed. A variety of standing meetings have also been set up to facilitate communications about transition issues. Nevertheless, there is no joint plan or document that describes measurements that are planned for transition, nor is there a process for creating them. In addition, ongoing changes in the space industry and meteorological community have introduced influences that could have long-term impacts on the approach to transitions. NASA experimented in the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) program by procuring ocean color data through a data purchase, and is exploring a similar approach in the Landsat Data Continuity Program to obtain land surface data. NESDIS has already purchased synthetic aperture radar (SAR) data from the Canadian Radarsat satellite for use in ice monitoring, and NASA has recently 13 Presentation by Michael Crison, NOAA’s National Environmental Satellite, Data, and Information Service, to the Committee on NASA-NOAA Transition from Research to Operations, June 2002, Washington, D.C. 14 Presentation by Sarah Horrigan, Office of Management and Budget, Science and Space Branch, to the Committee on NASA-NOAA Transition from Research to Operations, April 2002, Washington, D.C.
OCR for page 61
considered the development of a variety of SAR systems that could provide similar data. Opportunities for flying NOAA sensors on geostationary communications satellites, either as payloads-of-opportunity or as data buys, have also been explored. The continuously evolving diversity of users and user needs increases the demand for a faster transition process from R&D to operational applications. A process that provided for the quick and affordable transition of demonstrated technologies would respond to the “lessons learned” that are documented in the case studies in this report (Appendix B). Such a process can be facilitated by promoting the satellite observing system as adaptive and flexible so that R&D innovations can be quickly validated and then transferred into routine operational systems. In effective research-to-operations transition pathways, it is important that researchers have the resources (i.e., the “telescopes”) to allow them to be innovative. Research satellites are needed to test and validate revolutionary remote sensing ideas. However, DOD and NOAA want clearly defined and consistent observations that meet their known requirements. These contrary requirements of the research and operational user agencies need to be reconciled. One approach to satisfying both research and operational needs is to make every operational satellite capable of carrying advanced technology research instruments that can be space-validated in parallel with the operational mission. Once the advanced technologies are validated, the older and less capable satellite instruments can be retired from subsequent operational satellites, thereby making room for another round of new technology experiments. This approach (often supported by reserving some portion of the operational mission, e.g., 25 percent) would allow a steady infusion of new observing capabilities into operational satellites. Finally, coupled with the continuous validation of new instrument technologies, there is a need for applied research units at operational numerical prediction centers or associated test beds, which have the mission and capabilities to demonstrate the utility of the new observing technologies. In the concept of test beds, new data-assimilation and modeling algorithms are developed and evaluated prior to the launch of particular satellites and the arrival of their observations. Test beds can provide the infrastructure (computational facilities, numerical models, databases, and so on) for researchers to develop improved models and data-assimilation techniques and test the likely impact of future observations. Two test beds have recently been established in support of space weather and hurricane prediction (Boxes 5.1 and 5.2).
OCR for page 62
BOX 5.1 Transition from Research to Operations in Support of “Space Weather” Forecasting The transition from research to operations in the area of space weather and climate—that is, space meteorology—has many similarities to research-to-operations transitions in the area of weather and climate near Earth’s surface. However, space meteorology is significantly less mature than its near-Earth equivalent. The records are much less extensive, the sampling of the Sun-Earth system is much sparser, and many of the fundamental processes are yet to be understood. Until our understanding of space meteorology has matured to the point of being comparable to current understanding in meteorology, a successful early transition from research to operations will benefit not only from the lessons learned, findings, and recommendations in this report but also from particular attention to the following: Significant participation of researchers in all aspects of the transition process. Many of the links in the complex chain of processes between the solar interior and the near-Earth environment remain to be adequately explored. Close involvement of the scientific community in the transition process will be critical to ensuring early identification and evaluation of useful observables, early development of partial forecast models in preparation for comprehensive ones, and efficient use of resources. Stimulation of early development of partial forecast tools. Research models for space meteorology exist at present only for parts of the Sun-Earth system—for example, the National Center for Atmospheric Research community’s Thermospheric Ionospheric General Circulation Model.1 The development and validation of forecast models for parts of the system should be stimulated by efforts that run parallel to and interact with research projects. That development will provide early useful information to the user community and foster expertise and interest within the research community in the development of robust forecast tools. Use of instrumentation for both operations and research. In view of the enormity of the volume to be studied—spanning the Sun, the inner heliosphere, and all of geospace—the ensemble of research and operational instrumentation should be viewed as complementary in order to effectively increase the available research resources. Receiving input from both the research and user communities in the definition of instrumentation and having adequate funding to meet the joint requirements are important to catalyze progress. Other than these differences, the requirements to transfer proven, operationally needed space measurement capabilities from NASA research to NOAA operations and establish test beds for research models (e.g., the Community Coordinated Modeling Center at NASA’s Goddard Space Flight Center) are similar to the overall needs of the tropospheric weather support community (NRC, 2003). The National Space Weather Program has the transition of research to operations as a goal: “The overarching goal of the National Space Weather Program is to achieve an active, synergistic, interagency system to provide timely, accurate, and reliable space environment observations, specifications, and forecasts within the next 10 years.”2 1 More information on the Thermospheric Ionospheric General Circulation Model is available online at <http://www.hao.ucar.edu/public/research/tiso/tgcm/tgcm.html>. 2 Further background on the National Space Weather program is available online at <http://www.spacescience.org/SWOP/NSWP/>. For more information on research in space weather, see
OCR for page 63
National Research Council, The Sun to the Earth—and Beyond: A Decadal Research Strategy in Solar and Space Physics (NRC, 2003). That report provides a more detailed discussion of issues related to the transition from research to operations for “space weather” forecasting. Of particular interest is its recommendation that NOAA assume responsibility for the continuance and distribution of solar wind data from spacecraft at the L1 location as well as near Earth. See also the Web site <http://www.spacescience.org/SWOP/NSWP/1.html>. Accessed March 13, 2003. BOX 5.2 The Joint Hurricane Testbed The U.S. Weather Research Program (USWRP) is a coordinated effort among four agencies (NOAA, NSF, DOD, and NASA) that provides the essential focal point and organizational driver for moving ideas and technology from the research arena to operations. The USWRP’s Hurricane Landfall implementation plan proposed a feasibility demonstration in which “facilitators”—individuals designated to interact with researchers and the National Hurricane Center (NHC)—would transfer research to operations more efficiently and more effectively than is currently being achieved. The USWRP funding for the Joint Hurricane Testbed (JHT) began in Fiscal Year 2001. The JHT project funded about 15 research-to-operations projects during its first 2 years of existence. Several projects are already up and running at the NHC. The NHC Technical Support Branch has worked with the affiliated research groups selected to develop test bed projects to provide access to the databases, communications systems, and display systems at the NHC. However, the new USWRP investments in the transition of promising hurricane research into operations have required many adjustments at the NHC. As indicated from the first 2 years of the JHT test bed experience, the NHC can become overwhelmed by too many test bed projects. The saturation point is a function of the lack of NHC staff available to manage the test bed projects and the numbers and characteristics of the transition projects. The capabilities and availability of technical staff to handle the test bed projects have limited NHC’s ability to interact with researchers on project testing and evaluation. During the second year of the JHT operation, a JHT facilitator was added to the NHC staff to assist in managing the test bed projects, especially the requirements from the individual research teams for access to the NHC computing infrastructure. It is critical that the personnel requirements for these tasks—directing the efforts of the transitions and the ongoing operational evaluation support—be resolved to avoid any negative impact on the operational forecast and warning system. In summary, the JHT project is designed to expedite promising research on improving hurricane forecasts into the NHC operations and has infused important funding toward this issue, leading to initial success. The most important lesson emerging from the JHT project is—Don’t forget the end user in the transition process. In this case, the human resources at the NHC were not sufficient to receive and evaluate properly all of the incoming research projects. An effective distribution of the JHT funds among the transition participants, the NHC, and the researchers will help to ensure the best-possible result.
OCR for page 64
SUMMARY OF TRANSITION ISSUES NASA and NOAA share the motivation and need to transition research capabilities to operational status. Analyses of case studies indicate that effective transition pathways incorporate strong management, well-defined transition objectives and plans, effective processes for performing the transition, and adequate human and fiscal resources to accomplish the transition. Transitions have been largely implemented on an ad hoc, case-by-case basis rather than as part of an overall plan. While an effective relationship between NOAA/NESDIS and NASA/GSFC has been established for procurement and incremental upgrades of the POES (and now NPOESS) and GOES satellites, the relationship has been less successful in planning and implementing major system upgrades. The potential for NOAA to develop data-assimilation algorithms has been significantly increased by the establishment of the Joint Center for Satellite Data Assimilation. There is no well-defined agreement regarding which NASA systematic measurements will be transitioned to NOAA and which NOAA operational measurements require NASA research precursors. There is also no formal NOAA process in place to identify requirements for which NASA research measurements would be beneficial (although the committee is aware of efforts in NOAA toward the development of a strategic plan, including mechanisms to identify operational and policy requirements), and there is no NASA guidance for determining how research measurements should be prepared as candidates for operational transition. The mismatch between a NOAA planning process that requires a well-defined set of exploratory measurements and the unpredictability of the NASA research peer-review process is an impediment to this transition. Finally, there is no established program in either NOAA or NASA to infuse new technologies and to reduce technology risk for operational satellites, and there is no formal means for prioritizing transition activities other than on a case-by-case basis. The environment in which research-to-operations transitions occur is changing rapidly. Important influences include the increasing use of international partnerships, accelerating technological progress, the emergence of commercial data sources, and the increasing importance of numerical weather prediction and data assimilation using sensor data directly. The institutional and budgetary separation of NASA and NOAA makes effective research-to-operations transitions challenging. Effective transitions require close collaboration, planning based on mutual interests, jointly developed resources, and singular authority to make decisions that bind both parties. Neither NASA nor NOAA has the authority to establish plans that depend critically on decisions made in the other agency. With the current organizational constraints, the research-to-
OCR for page 65
operations transition is likely to continue to be characterized as “passing issues over the wall,” with operational needs and research capabilities planned and executed largely independently of each other. Thus, although NASA and NOAA desire a smooth transition of research to operations, the current environment is characterized by the lack of an overarching mechanism to ensure that transitions benefit from a common process and are, in general, efficient and effective.
Representative terms from entire chapter: