Informing Decisions in a Changing Climate (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 A National Initiative for Decision Support A s the impacts of climate change become apparent, people are begin- ning to realize that some of their decisions should take account of the changing climate. At least some of them have not previously thought of a changing climate as a factor relevant to the institutional choices for which they are responsible. Most of these climate-affected decision makers throughout the country need decision support. Climate change means that the future—sometimes, the near future—is going to produce situations for which the nation is currently unprepared. Standard practices—in fields from regional planning to infrastructure management to public health to emergency preparedness—that worked well in the past under an assumption of a stationary climate are no longer appropriate. Human institutions will need to change the processes by which they inter- pret and use climate-related knowledge. They will also need new kinds of information to help them change. And the change will need to be continual because climate will continue to change, sometimes in a foreseeable way and sometimes with surprises. Decision support poses major challenges to decision makers and social institutions. The challenges include developing new approaches to com- munication between scientists and decision makers, adopting new ways of thinking and planning to incorporate longer time horizons and the major uncertainties of climate change, evolving new modes of social learning, and developing previously underdeveloped areas of science. Meeting these chal- lenges will require a process of social learning at a large scale that occurs over decades, and its results cannot now be envisioned clearly. 125 

126 informing decisions in a changing climate Societal responses to industrialization—a fundamental change of the twentieth century—suggest the important role decision support can play when society faces major changes. Industrialization prompted the inven- tion of new institutions—including labor unions and social security—and new decision support systems, including national economic statistics on inflation, unemployment, the gross domestic product, and the like. Those systems have become essential to the emergence of macroeconomic policies in the United States and other countries since the Depression (Stein, 1996; Hall, 1989). The indicators—together with the organizations that devise, monitor, validate, archive, and interpret them, the annual reports of the Council of Economic Advisers in the executive office of the President, and the debates they shape—are decision support systems for national economic policy. They have contributed to development of a common conceptual framework and a set of national and international institutions and have become indispensible to economic decisions in the public and private sec- tors. A lesson of this history is that decision support can make a difference when the federal government provides leadership in data gathering and analysis—even in that most decentralized of activities, a market economy. Climate change will require a similarly wide-ranging process of social learning. It potentially affects all sectors of economic activity; all regions of the country; all levels of governmental and social organization; and a great variety of professions, communities, and individuals. It is useful to think of those needing climate-related decision support as constituen- cies—collections of decision makers that may be defined by the kinds of climate-related hazards or opportunities they face, the kinds of climate-af- fected decisions they must make, shared legal or regulatory mandates, a regional location, or the fact that they are already organized as a constitu- ency. Focusing decision support efforts on constituencies is an effective way to organize them around users’ needs. Some climate-affected constituencies are linked to government agencies with mandates to assist in their decision making—including local air and water quality agencies and the Air and Water Offices of the Environmental Protection Agency (EPA); coastal and water managers and the Climate Program Office of the National Oceanic and Atmospheric Administration (NOAA); emergency responders and the Federal Emergency Management Agency—and some are not. Some of those affected already have formal organizations that can collectively request and receive decision support; others do not. The key principles for implementa- tion of decision support (see Chapter 2) are well established from research and experience in operations research and the decision sciences and in vari- ous endeavors to make scientific knowledge information useful for practical decision making, and apply across constituencies. The principle of designing for learning is particularly important over the long run. As the climate is changing, change is also occurring in scien- 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 127 tific understanding, in experience with response options, and in the needs identified as critical for decision making. In such an environment, effective decision support cannot be reduced to a stable bureaucratic formula or be optimally designed from the outset. Rather, climate decision support must be designed to recognize local contexts and surprises and to improve over time, as climate change unfolds, as more organizations across society come to address climate change on a regular basis, and as scientific understanding of climate changes. Several types of deliberate learning processes, including program evaluation and adaptive management, could prove effective. But an analytic-deliberative approach to decision making and learning, which integrates scientific information into a broadly participatory and iterative process of appraisal and reconsideration, is usually best suited to the kind of decision environment that is typical in responding to climate change: one characterized by changing physical conditions, changing information, and multiple participants with different and sometimes changing objectives. THE FEDERAL ROLE IN DECISION SUPPORT It is important to emphasize that in developing new structures and in- stitutions for decision support, the federal government plays an important but not exclusive role. To change the energy system and learn to live with a climate system that is no longer stationary, millions of decision makers— state and local governments and their agencies, large and small businesses, nonprofit organizations, as well as individuals and households—will need information, and much of this information will be provided from sources other than the federal government. This multiplicity of information sources is desirable because there will be no one best source of information for the wide range and variety of decision makers with their varied needs. A federal role and federal leadership are essential, however, for inform- ing national responses to climate change. In Chapter 1 we identify four roles for federal agencies: (1) service to agencies’ constituencies and to other climate-affected constituencies that cannot otherwise get the information they need; (2) international collaboration; (3) provision of public goods (research, observations, communications links, etc.); and (4) facilitation of decision-making processes by nonfederal entities. Chapters 2, 3, and 4 in- clude several recommendations regarding ways to perform these roles. Decision support requires resources: money, new kinds of expertise, training of people in new skills, new interorganizational relationships, and supportive forums and organizations. The federal government is not the only source of these resources. Businesses will find opportunities to profit by supplying decision support in the form of new services and products, as already occurs, for example, with consulting firms and with companies that highlight the need for—and then offer to sell—carbon offsets for air travel- 

128 informing decisions in a changing climate ers. Professional associations, such as the Association of State Floodplain Managers, and philanthropic organizations also provide resources. For example, the Climate Works Foundation, formed with support from several major philanthropies, aims to assist India, China, the United States, Latin America, and Europe to undertake large-scale mitigation of greenhouse gas emissions. Communities at many scales, from Alaska native settlements facing melting permafrost to professional associations revising codes of practice to take account of changing climate, are already generating deci- sion support for their members—with potential lessons for others. And of course, public and private scientific institutions, which were the first to rec- ognize climate change, will also contribute. Still, none of these institutions and organizations can take the place of the federal government. A NATIONAL INITIATIVE Federal efforts to provide and promote climate-related decision support should be coordinated within a new integrated, interagency initiative. All the recommendations for federal action described in the previous chapters can be implemented in an integrated fashion under the umbrella of the initiative, as described below. Recommendation 9: The federal government should undertake a na- tional initiative for climate-related decision support under the mandate of the U.S. Global Change Research Act (USGCRA) and other existing legal authority. This initiative should include a service element to sup- port and catalyze processes to inform climate-related decisions and a research element to develop the science of climate response to inform climate-related decisions and to promote systematic improvement of decision support processes and products in all relevant sectors of U.S. society and, indeed, around the world. The Service Element in the National Initiative The service element of the national initiative should support the cre- ation and expansion of decision support networks and processes that imple- ment the principles of effective climate-related decision support in federal agencies and beyond. To do so, it should support demonstration and de- velopment activities to promote the emergence of decision support systems to serve climate-sensitive constituencies that are currently underserved. Priority should be given to constituencies on the basis of agencies’ man- dated responsibilities, the vulnerability of their mandated and potential constituencies to climate change, and the difficulty of those constituencies in developing resources for themselves. These activities might include new or 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 129 expanded programs to serve agencies’ constituencies, support for decision dialogues, pilot programs to demonstrate effective processes for climate de- cision support in particular sectors, and other constituency-based activities to define decision support needs and build communication networks. The national initiative should also support learning networks that link multiple decision support activities and thus facilitate informal learning across the activities and support state and local governments in develop- ing decision support services with and for them and their constituencies, following the federal model if they have not already developed their own. Recommendations 2 (in Chapter 2) and 5 (in Chapter 3) offer details. The initiative should also build national decision support capacity by making investments in human resources and in institutionalizing practices that consider climate change in decision making. For example, federal support will be needed for training people to produce and use knowledge, methods, databases, and indicators inspired by the need for climate-related decision support and to improve communication between the producers and users of decision support products (see Recommendation 8 in Chapter 4 and the discussion of the research element below). The national initiative and participating agencies can develop the na- tional capacity to identify climate-sensitive constituencies and interact with them by providing human resources, funds, professional training and educa- tion, and leadership in federal environmental and climate agencies, encour- aging a systematic perspective, and in some instances establishing mandates to use climate-relevant information. Moreover, federal agencies can work with climate-sensitive constituencies to develop draft rules, standards, and procedures for the appropriate use of climate-related information for mak- ing decisions, such as for infrastructure construction in low-lying coastal areas and water and fire management in areas expected to see increasingly severe droughts. Agencies can also support pilot-scale demonstrations of such new standards and procedures. Federal agencies that serve constituencies that need climate-related decision support should provide such services, following the principles of effective decision support process and using the model of deliberation with analysis where feasible. These include agencies now participating in the Climate Change Science Program (CCSP), but also many others. As noted above, federal agencies can also encourage and facilitate the provision of decision support by state and local governments and nongovernmental or- ganizations to nonfederal constituencies around the nation. The Regional Integrated Science and Applications (RISA) Program centers of NOAA provide operational examples that show how a government agency can facilitate the development of effective decision support processes. The service element of the national initiative should include the peri- odic assessment required under Section 106 of the USGCRA, to integrate 

130 informing decisions in a changing climate and interpret scientific findings and uncertainties related to the effects of global change and current and projected major trends in global change. In discharging this legislative mandate, the Global Change Research Program should follow the principles of effective decision support: In particular, the assessment should flow from a participatory process that is driven by users’ needs. In recent years, a series of synthesis and assessment products has been put forward as a response to the requirement of Section 106. Al- though several of these products and the “Unified Synthesis Product” now in preparation provide useful information for decision making, a systematic user-driven process should underlie the development of these assessments in the future to increase the usefulness and use of the information they provide. The Research Element in the National Initiative The research element of the national initiative should develop the sci- ence of climate response with two foci: research for decision support, aimed at providing decision-relevant knowledge and information; and research on decision support, aimed at making decision support activities and associ- ated decision processes more effective, efficient, and capable of learning. A major new scientific research effort is needed to better inform re- sponses to climate change. Attention to the needs of the users of scientific information—a cornerstone of decision support process—will help generate some of the new scientific research priorities. As noted above, this effort needs to draw on many fields of scientific inquiry, scholarship, and analysis, of which climate science is only one. The effort would particularly include the social, behavioral, and economic sciences and integrate them with re- search, concepts, and methods from the natural sciences and engineering. To provide the needed research for informing climate change response, the national initiative should expand the current federal research effort from its current focus on climate processes and technology. That expansion would add research on changes in society that determine the trajectory of climate change and shape its impacts, about the ability of people and orga- nizations to respond, about effective ways to respond, and about the likely intended and unintended consequences of the various responses people are considering. A major expansion of research on these topics is already required simply to meet the needs identified by the existing constituencies of NOAA and EPA—and the demand is likely to expand both in substance and in breadth as climate changes continue. In Chapter 4 we identify some general rubrics under which much of the needed research is likely to fall (see Recommendation 6). However, the research program for decision support cannot be fully defined in ad- vance because it will emerge in part from dialogue between those who need decision support and those who will produce it. Such discussions can 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 131 be expected to generate research questions that are not yet on anyone’s agenda. For example, dialogue will often identify needs for knowledge and information focused on very specific questions of importance to particular decision makers. Achieving the objectives of the research focus on decision support will require investments in monitoring and evaluation of decision sup- port activities, processes, and products and the institutional resources and infrastructure to support analysis, deliberation, and iterative learning. To achieve these objectives, the national initiative should include support for a clearinghouse function—either a formal organization or a set of distributed organizations and activities (e.g., research centers, websites, interactive databases, Wikis) that collect information on the results of decision support efforts, conduct research on these results, and make the knowledge gained widely available to facilitate learning by future decision support activities. Such organizations and activities would gather both formal evaluations of decision support efforts and case histories and other qualitative information on these efforts and make them accessible so that those developing new decision support activities and products can learn from the experiences of others. The national initiative should provide leadership in developing ways to assure the quality of information provided by those organizations and activities. Some of the research recommended in this report would use those organizations and activities both as sources of data and as disseminators in making research results broadly available. Observations and Human Resources in the National Initiative The national initiative should also expand observational systems to provide information needed for decision support and strengthen the work- force needed to conduct the recommended research and better link the providers and users of information for climate-affected decisions; see Rec- ommendations 7 and 8 (in Chapter 4). The expansion of observational systems will link existing data on physical, ecological, social, economic, and health variables relevant for climate-related decision support from many sources and develop new data as needed. Important efforts in this expansion will include improving accessibility of data at relevant scales, developing effective ways to link data of different kinds, developing useful indicators from the data, and linking U.S. data systems with international data development efforts. ORGANIZATION OF THE INITIATIVE Although the idea of a national initiative could be interpreted as imply- ing a new organizational entity, we do not think it advisable to centralize the initiative in a single agency. Doing so would disrupt existing relation- 

132 informing decisions in a changing climate ships between agencies and their constituencies and formalize a separa- tion between the emerging science of climate response and fundamental research on climate and the associated biological, social, and economic phenomena. However, our recommendations do imply significant change in the ways many federal agencies serve their constituencies, coordinate with each other and with nonfederal decision makers, and set priorities for research. They also identify an important new federal responsibility—to facilitate distributed responses to climate change—that has no obvious agency home. And they call for significant changes in the nation’s research program related to climate change, including the development of scientific fields that have received little support in the past, that have not recently been mission priorities of any federal agency, and that the agencies are ill prepared to develop. A major organizational effort will be needed to ensure that the recom- mended activities form a coherent initiative in which the whole will be more than the sum of the parts. Significant restructuring from the current structure of the CCSP and the U.S. Climate Change Technology Program (CCTP) with their interagency working groups will be required. Most of the elements of the initiative will still require an interagency coordination structure, but its form will be different from the current one (see below). The most effective organizational form for implementing our recom- mendations is best left to the new national administration to decide, as it may simultaneously be reorganizing other activities of the relevant agencies. We note, however, that we do not believe any new legislative authorization is needed. The necessary coordination can be achieved under the USGCRA of 1990, which provides the necessary responsibility to the National Science and Technology Council (NSTC). The NSTC can develop the recommended initiative in coordination with the new climate and energy office in the Executive Office of the President. The following sections discuss the main issues for restructuring and offer our suggestions for them. More Participating Agencies Understanding responses to climate change and serving climate-af- fected constituencies will require the involvement of many more agencies than now participate in the CCSP, as well as of offices that are in CCSP member agencies but are not now part of the program. For example, land and water management agencies, such as the Bureau of Land Management, the National Park Service, and the Bureau of Reclamation, are affected by climate change, need decision support, and would benefit from being bet- ter integrated with climate research. Agencies that produce data on human activities that drive or are affected by climate change, such as the Census Bureau, the Bureau of Economic Analysis, and the Energy Information 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 133 Administration, also need to be involved in the initiative. Agencies that sup- port science education should also participate. With such an increase in the number of organizations and a proliferation of tasks—including providing services, doing research, and developing observation systems—the existing interagency working group organization will require modification. For ex- ample, a special interagency working group might be needed to coordinate the development of data systems to link environmental and socioeconomic data. New Tasks It makes sense to organize the decision support efforts for particular constituencies in the agencies that serve those constituencies, as is done by the NOAA Sectoral Applications Research Program for coastal and water managers. But, in many cases, the most obvious agencies do not consider such decision support activities as part of their missions and lack offices and personnel with the responsibilities and expertise needed to manage the research. For example, it would make sense to locate decision support research and services related to energy-efficient choices in homes in the De- partment of Energy (DOE). However, such activities have been peripheral to the DOE mission for quite some time, and the department has not sup- ported research on household energy choices since the early 1980s. Many environmental agencies, like DOE, define their research mis- sions mainly in terms of studying environmental phenomena as physical, biological, and engineering questions, not in terms of studying human– environment interactions. With such foci, human–environment research is neglected in terms of budgets, the available staff may not have the skills or training to manage the efforts, and this research area may not receive the requisite support from higher levels in the agency. There is a lack of inte- gration across bodies of knowledge and practice needed for effective deci- sion support throughout the federal government, as shown in the analyses presented in Chapter 2. Thus, in responding to demands for decision support, officials responsi- ble for the national research effort will need to find ways to ensure that the relevant agencies have the needed funds and staff and that agency managers are given the responsibilities and resources needed for the programs. The lack of expertise and funds is especially acute for the social sciences: many environmental agencies have historically lacked staff with the requisite ex- pertise, funds, and organizational commitment in those fields. Some of the new tasks in the decision support initiative do not have obvious homes in the federal government. One of these is research on c ­ limate-related decision support. Research on decision support for par- ticular constituencies can usefully be decentralized to mission agencies, but 

134 informing decisions in a changing climate research on decision support processes and methods of general applicability has no obvious home in any environmental agency. The National Science Foundation (NSF) does and can support some such research through its Division of Behavioral, Social, and Economic Sciences, which supports basic research on communication and decision making. However, still lack- ing is a place for enterprises that would link basic research studies to envi- ronmental and socioeconomic observations, to workforce development in decision support science and services, or to the federal agencies and other organizations that provide decision support. NSF might serve as an ap- propriate coordinating point. However, because these linking functions do not currently fit comfortably within the central mission of NSF, leadership and skillful coordination would be needed by the agency and its partners to make scientific research on decision support useful. Relationship to Ongoing Research The decision support initiative we recommend would complement the existing efforts of the federal research program (the CCSP and the CCTP) by adding the foci of research for decision support and on deci- sion support. We believe that the research program needs expansion in other ways as well, in order to develop fundamental knowledge in areas that provide the scientific underpinnings of research for decision support (such as research on the social drivers of energy consumption, land-use change, and climate vulnerability; on decision making under uncertainty; and on institutional mechanisms for controlling resource use; see Stern and Wilbanks, 2008). A new report from the Committee on Strategic Advice to the U.S. Climate Change Science Program (National Research Council, 2009b) pro- vides recommendations on future priorities for the CCSP. We believe that the research recommendations of that study and ours are mutually support- ive. The Strategic Advice Committee report addresses fundamental research issues that are beyond the scope of this report. However, its advice on ways to expand fundamental knowledge to inform decision making is consistent with our recommendations for the research element of the national deci- sion support initiative. Together, the two reports call for significant change in research activities being conducted under the authority of the USGCRA, including developing underdeveloped areas of research and finding appro- priate organizational homes in the federal research program for research areas that do not now have them. Both reports also call for development of research along a continuum from fundamental to narrowly applied, but linked together by their relevance to understanding and responding to climate change. 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 135 Central Authority and Collaboration The current CCSP office has a coordinating function and as such, has a minimal operational budget. It has no direct authority over the partici- pating agencies’ climate science research agendas or related programmatic activities. Continuation of this situation is likely to pose a serious roadblock for expanding the research program and adding a service element, especially because the participating agencies will likely face considerable budgetary pressures. There are several possible ways to address this challenge, such as providing a central budget to the coordinating office or offices and using the national budgetary process as described in the USGCRA (through the Office of Management and Budget) to allocate needed funds to the relevant agencies. Because some of the new tasks entail change in agencies’ tradi- tional missions, effective leadership by the new White House coordinator of energy and climate policy will be needed to ensure that any distributed funds are used for the national initiative. It seems likely that successful implementation of the initiative will also require high-level commitment in the participating agencies to the new decision support mission. The federal effort on climate-related decision support will have to involve unusually effective collaboration among federal agencies because many agencies need decision support, provide information needed for deci- sion support, or serve constituencies that need decision support. Thus, the effort should extend far beyond the agencies currently involved in the CCSP. This breadth will require strong leadership from the Executive Office of the President, including the science adviser and the new coordinator of energy and climate policy. To fully meet the mandate under the USGCRA to “assist the Nation and the world to . . . respond to human-induced and natural processes of global change,” we believe that the federal government, through the NSTC, will need to comprehensively reformulate its plan for implementation of the act. The next phase of that implementation should include research on the fundamental biophysical and human dimensions of climate change, as well as research for and on decision support, improvement of observational sys- tem and indicators, and monitoring, service provision, and human resource development. The research effort should expand on the activities currently being pursued under the CCSP and the CCTP and be closely integrated with those research activities. Any federal activities that might be included in a national climate service (see below) should be part of the initiative. The current climate research enterprise is organized under two inter- agency working groups: the CCSP and the CCTP, with other interagency groups functioning under them. We believe a similar general structure will be appropriate in the future, although with the groups defined differently. Organizing by climate-related societal issues, such as suggested by the 

136 informing decisions in a changing climate Committee on Strategic Advice on the CCSP (National Research Council, 2009b), is an attractive possibility. Within that recommended organization, some decision support research and service activities could be organized topically, but others, such as development of observational systems and human resources and research on decision support processes and methods, might require a cross-cutting organization. Such a structure could help coordinate research and development on technology with research on its costs, benefits, adoption, and use—an important coordination need for climate response. Relationship to a National Climate Service The idea of a national climate service in or led by NOAA has received considerable attention in recent years. As of this writing, the idea is in flux, with no agreement on its purview, mandate, or organizational location. In our view, any form of national climate service should implement the prin- ciples of effective decision support. Thus, it should develop decision support products by means of communication between information providers and users. If this communication is productive, scientists will alter some of their research priorities to make their research more useful. For this reason, we believe any national climate service should be part of our recommended decision support initiative and linked to its research element. There are different views regarding whether decision support activities, including a National Climate Service, should be separate from or integral to the research program. For example, the Committee on Strategic Advice on the CCSP (National Research Council, 2009b) recognizes the need for strong linkages between the research and service components, but raises the argument that demands for service could overwhelm the research program. Because of the evidence that decision support is most effective when it emerges from a process that is guided by users and that such a process can shape research agendas in ways that yield more useful research products, this panel concludes that close coordination between service and research functions is critical for making research useful for decision support. More- over, we believe the lesson from past experience is that research and service functions both do better when they collaborate than when they proceed separately. One concept of a national climate service is that a new organizational entity would be created in NOAA, modeled on the Weather Service, that would transform data from climate models into decision support products intended for use by various kinds of decision makers and would disseminate that information publicly. We do not believe this model by itself would meet national needs for climate-related decision support because it does not fully implement the principles of effective decision support. In particular, this 

A NATIONAL INITIATIVE FOR DECISION SUPPORT 137 model does not make adequate room for communication with an input from information users, especially those who do not normally interact with NOAA. Such a climate service would not be user driven and so would likely fall short in providing needed information, identifying and meeting critical needs for research for and on decision support, and in adapting adequately to changing information needs. In addition, this model of a climate service is focused only on providing information about climate: It would therefore fail to develop and provide the many kinds of nonclimate information that climate-affected decision makers also need. If a national climate service is created, we believe it should follow a much more user-driven and interagency organizational model, be closely linked to the research program, and have a purview that goes beyond de- veloping and providing information about climate. We also believe that in addition to any new organizational entity, such as a climate service, indi- vidual federal agencies should develop efforts to provide decision support for their climate-affected constituencies. The realization throughout the nation that Earth’s climate is changing frames a moment of need and opportunity. The need, emerging over the years ahead, is for knowledge to inform Americans about the implications of the changing climate in their personal choices, organizational responsi- bilities, and public policies. Growing out of the strong base of analysis in the natural and applied sciences of weather and climate forecasting and with additional investment in the science of climate response, there is an opportunity to empower people to face a transition to a world that people have remade and continue to remake.   This text was changed from the prepublication version to clarify the panel’s meaning.