National Academies Press: OpenBook

Informing Decisions in a Changing Climate (2009)

Chapter: 3 Decision Support and Learning

« Previous: 2 Effective Decision Support: Definitions, Principles, and Implementation
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 71
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 72
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 73
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 74
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 75
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 76
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 77
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 78
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 79
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 80
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 81
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 82
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 83
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 84
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 85
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 86
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 87
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 88
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 89
Suggested Citation:"3 Decision Support and Learning." National Research Council. 2009. Informing Decisions in a Changing Climate. Washington, DC: The National Academies Press. doi: 10.17226/12626.
×
Page 90

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.

3 Decision Support and Learning A changing climate presents two major challenges to decision mak- ers and to those who provide decision support. As discussed in Chapter 1, some decisions and decision-making routines will need to change to reflect the changing climate. Also, since climate change and its interactions with society are dynamic, the rules for making those decisions will need to continue to change over time. That is, decision makers must not only change, but be prepared to continue changing. In a sense, decision makers will need to plan to be surprised. The systems that support climate-related decisions must become adap- tive systems, learning through a variety of means. Some decision support systems are already adapting in this manner. For example, the climate deci- sion support system for New York City has evolved as scientists, govern- ment officials, and activists develop working relationships to tackle some problems and transform their understanding of the situation they face; see Appendix A. This chapter first addresses the challenges of learning in the context of climate-related decision support. We next consider four modes of learning and explain why one, which we call deliberation with analysis, is the most appropriate for meeting the challenges of response to climate change. We then recommend ways that the federal government can apply this mode of learning in its own decision support activities and facilitate decision makers around the country in making adaptive responses to climate change and learning from their own and others’ experiences. 71

72 informing decisions in a changing climate CHALLENGES Climate change and many other environmentally related policy prob- lems are members of a class of “wicked problems” (Rittel and Webber, 1973)—problems with no definite formulation and no clear point at which the problem is solved. They have been described as having five key charac- teristics (see Dietz and Stern, 1998): 1. Multidimensionality: A single environmental process or policy can have many different types of effects, distributed unevenly so that those af- fected face unequal shares of the costs, risks, and benefits. 2. Scientific uncertainty: Current understanding is primitive in com- parison with what decision makers want to know—and sometimes the de- gree of uncertainty is itself uncertain. In addition, the consequences unfold at an unfamiliar tempo, with some effects delayed and others disconcert- ingly prompt. 3. Value conflict and uncertainty: People differ in the importance they attach to the different effects of any action, and these judgments change as people experience how their own and others’ actions affect the things they value. 4. Mistrust: Decision makers are often mistrusted by those their deci- sions affect; their analyses are also often mistrusted. 5. Urgency: It is not feasible to postpone action until scientific uncer- tainties are resolved. In addition to these characteristics, climate change presents a dynamic decision context and unfolds over a time scale that extends beyond the planning horizons of most organizations and over a geographic scale that exceeds their control. Learning by doing under such conditions creates challenges for leader- ship. Although it makes sense to treat all decisions as provisional, such an approach is not easily reconciled with conventional notions of accountabil- ity. Decision makers will have to discard well-accepted standard procedures that offer them a kind of protection in favor of new ones that may be more effective, but that will open them to criticism when, inevitably, errors occur. Another challenge is that most climate change decisions will be undertaken in a decentralized fashion, as local and state governments, firms, and other institutions respond to a changing climate. Thus, the federal role in decision support will have to be aimed at creating and informing a distributed capacity to make sensible choices. This is both functionally necessary and advantageous to the nation as a whole, since decentralized decision making will generally be better able to cope with surprises and specific local conditions. Nevertheless, federal

DECISION SUPPORT AND LEARNING 73 agencies should also be prepared to address issues that arise repeatedly in multiple localities or sectors and might therefore benefit from national-level attention. LEARNING MODES The panel examined four kinds of learning in organizations in terms of their suitability for meeting these challenges. As Table 3-1 indicates, these four modes of learning span a range of assumptions about the context and processes of decision making. We analyze the modes of learning in relation to the main challenges of decision support in a changing climate. 1. Unplanned learning is a default mode: actions are undertaken without any explicit consideration of learning, and any change that occurs is unplanned and often unbidden. 2. Program evaluation involves formal assessment, often by outside parties, of a program’s effectiveness, with the expectation that adjustments will be made in response. 3. In adaptive management, actions are designed as experiments so that they will perturb the decision environment and thereby generate infor- mation useful for future adjustment and improvement. 4. Deliberation with analysis is an iterative process that begins with the many participants to a decision working together to define its objec- tives and other parameters, working with experts to generate and interpret decision-relevant information, and then revisiting the objectives and choices based on that information. Each mode of learning offers different strengths and weaknesses, and there is insufficient evidence to draw definitive conclusions about which mode is best for which situation. Nonetheless, the panel judges that de- liberation with analysis provides the learning mode best suited to a wide range of climate-related decision support applications. We note, however, that deliberation with analysis is not easily implemented. The rest of this section discusses each mode of learning. Unplanned Learning As has long been recognized by researchers (e.g., Cyert and March, 1963; Lindblom, 1959; Kingdon, 1984), much learning in organizations is unplanned. An organization may respond to events as they occur, but it devotes little attention or resources to making the learning process more effective. Unplanned learning may be attractive because it imposes no immediate costs in staff time and financial resources. It also weakens ac-

74 informing decisions in a changing climate TABLE 3-1  Modes of Learning LEARNING MODES Program Adaptive Deliberation Characteristics Unplanned Evaluation Management with Analysis Assumed Stable Stable Changing Changing decision environment Assumed Unitary Unitary Unitary Diverse decision maker Goals Implicit Set by decision Set by decision Emerge from   maker   maker   collaboration Stable Stable Potentially changing Data for Unsystemic Explicit indicators Explicit Explicit indicators learning Evaluation at end   indicators Continual Continual   monitoring   monitoring Means of Ad hoc Formal assessment Formal or Formal assessment appraisal Usually summative   informal   with deliberation Continuing   on its import Continuing Incorporation Unplanned Adjust after Continual Continual of learning   evaluation   complete countability and makes shortcomings and errors hard to identify. Such at- tractions are understandable in a decision environment that makes errors likely. (Unplanned learning is not the same as deliberate trial and error or adaptive management, described below.) However, the underlying assumptions of unplanned learning—that the decision environment is stable and the decision maker is unitary—do not fit at all well with the decision environment created by climate change. With no systematic attempt to make goals explicit or monitor performance, an organization may persist in ineffectual activities and fail to respond ef- fectively to change. Eventually, there are likely to be failures that are large and readily apparent. Program Evaluation Program evaluation has become a well-established field of applied so- cial science and professional practice, as well as the most familiar means of

DECISION SUPPORT AND LEARNING 75 formal learning in large organizations (see, e.g., Russ-Eft and Preskill, 2001; Chelimsky and Shadish, 1997; Shadish et al., 1995; Weiss, 1972, 1998). As defined by Mark et al. (2000:vii), evaluation helps people, individually and collectively, make sense of policies and programs “by providing systematic information about such things as the outcomes or valued effects of a social program, the cause of program success or failure, and the degree to which policy directives are being followed.” To learn from program evaluation, decision makers need to identify ex- plicit goals for a program, develop indicators of performance towards those goals, and gather data on the indicators. Evaluators compare post-program indicators with preprogram measures or with a scenario or situation in which the program was not adopted and an alternative course of action or no action was implemented (Newig, 2007; Rowe and Frewer, 2000, 2004). Program evaluation for climate-related decision support might make assess- ments at some designated end point or at each stage of the policy process. In either case, the evaluation might lead to adjustments in the budget, staffing, or other aspects of the program. Standard program evaluation presumes a stable decision environment and clear, stable goals. It has value for assessing climate-related decision support, although there are practical challenges: diverse participants in the decision may have different goals, and processes as well as outputs require evaluation (Moser, in press). If a decision support effort aims to help reduce vulnerability to drought of a county’s agriculture, outcome measures for the underlying components of vulnerability, such as exposure, sensitivity, and coping capacity or resilience, are required (e.g., Adger, 2006; Cutter, 1996; Luers et al., 2003; Schröter, Polsky, and Patt, 2005; Turner et al., 2003). If a goal of decision support is to change decision makers’ understandings of the importance of climate change to their operations, program evalua- tion must assess the content and quality of dialogue, the types of questions asked, and the level of concern and interest, since all of these may be rel- evant indicators. Such processes and outcomes may be measured in many ways (see Morgan et al., 2005; Moser, 2005b; Shackley and Deanwood, 2002; Tribbia and Moser, 2008). Program evaluation has proven valuable for strengthening accountabil- ity in government programs. It can be an effective framework for support- ing learning and improvement in programs, particularly when a program’s goals can be clearly defined, there is a single decision maker (or organiza- tion) with clear responsibility and authority to achieve those goals, and a relatively unambiguous connection can be made between observable data and the organization’s progress toward those goals. Unfortunately, the con- ditions for good program evaluation do not characterize many applications of climate-related decision support. Climate-related decision support may often occur in novel, poorly understood, and changing circumstances, with

76 informing decisions in a changing climate multiple decision makers pursuing multiple goals. Moreover, the connec- tion between measurable indicators and an organization’s progress towards goals may remain ambiguous and a subject of contention among the parties to the decision. Conventional program evaluation offers no means to re- solve such ambiguities, although a practice of “developmental evaluation” (Patton, 1994, 2007) is gaining adherents. Developmental evaluation puts the evaluator into a role of facilitating the process that we call deliberation with analysis (see below). Adaptive Management Adaptive management is a mode of learning intended for situations in which decision makers have a poor a priori understanding of the connection between their actions and their goals and therefore have much to gain if they learn by doing (Holling, 1978, 1996; Ludwig, 1996; Walters, 1986). A central argument of adaptive management theory is that learning from policies uncovers uncertainties and improves managers’ ability over time to respond to inevitable environmental, social, or economic surprises. Adap- tive management theory calls for policy interventions to be treated explicitly as experiments: carefully planned and monitored with replication and com- parison of management treatments (or lack of treatments) at appropriate spatial and temporal scales. Rather than presuming that managers make one-time decisions on the basis of the best existing knowledge, adaptive management regards policy choices for complex environmental problems as part of a carefully planned, iterative, and sequential series that emphasizes monitoring and learning as the system changes, both in response to exter- nal stimuli and in response to managers’ actions (Walters, 1986). Adaptive management differs from conventional management models in its explicit emphasis on iteration. Adaptive management embraces potential failures as data that provide opportunities for learning and the basis for better future decisions. For obvious reasons, however, this double-edged sword of “successful failures” has served as an institutional, political, and emotional barrier to the imple- mentation of adaptive management (Lee, 1993, 1999). Adaptive management presumes that a policy intervention, such as decision support, operates in a changing environment and that it might perturb that environment. In this respect, it is well suited to the decision environments presented by climate change. However, few adaptive manage- ment efforts have approached the ideal of iterated policy experimentation. It is difficult in practice to provide a control case in which a policy interven- tion that was believed to be beneficial was withheld. Local political interests often prevent the adoption or implementation of ideal experimental designs, for example, because of reluctance to accept the role of control group.

DECISION SUPPORT AND LEARNING 77 Moreover, because of the cost of complex experimental designs, the strong practical incentives against documenting failure, which lead to a tendency to design implementations that do not have adequate statistical power, make it difficult to be sure whether the policy made a difference. In climate-related decision support, adaptive management may be dif- ficult to implement because goals are diverse, outcomes delayed or hard to measure, and the relationships needed to manage the experiment are fragile. Gregory et al. (2006) thus suggest that adaptive management be adopted cautiously. They identify four clusters of conditions under which they hy- pothesize that those involved are likely to find it useful: 1. Spatial and temporal scale: Adaptive management is most easily im- plemented on relatively small scales that allow for management control. 2. Dimensions of uncertainty: Adaptive management is more likely to be considered worth the cost if, given the uncertainties in the process and the time available for learning, an experimental approach can produce results that are clearly interpretable to decision makers. 3. Costs, benefits, risks: Adaptive management designs are more likely to be considered useful if they include rules for stopping and chang- ing course which can keep the risks to all stakeholders at an acceptable level. 4. Institutional support: Adaptive management is more likely to be accepted if the participating institutions and affected groups have good leadership, the capacity to design and execute adaptive management, and the flexibility to learn, adjust, and avoid unacceptable tradeoffs. Given all the conditions that must be met, explicit experimentation is rarely practical in climate change applications. The field thus distinguishes active adaptive management, in which policy actions are explicitly designed to help generate learning as well as achieve program goals, from passive adaptive management, in which information collection is explicitly designed to improve the prospects of reliable inference from observing the effects of policy actions taken solely to achieve program goals. Arvai et al. (2006a) have argued that passive adaptive management is an important element of the decision support provided by the Intergov- ernmental Panel on Climate Change (IPCC). Many different management actions are now being undertaken globally by the multitude of sovereign political actors, private organizations, and institutions responding to chang- ing climates. Spatial variations in economic, social, and climatic conditions and in policies provide the potential for a database with variation on mul- tiple decision-relevant factors. Although these activities have developed unintentionally, they can provide an important source of observational data

78 informing decisions in a changing climate for climate policy that could be the basis for an intentional, international effort to learn. Ongoing decision support efforts can generate a similarly useful da- tabase for learning how to make decision support more effective. How- ever, the necessary information network does not exist to track, measure, monitor, and interpret the results of those experiences—especially with geographically dispersed and vulnerable groups. For adaptive learning, it is important that decision support initiatives have resources for data collec- tion and to develop new understandings from the experiences. Adaptive management (passive or active) can be a powerful tool for learning. The conceptual apparatus is well developed, together with specific ideas about implementing an adaptive approach (see Margolis and Salafsky, 1998). These ideas have been adopted by several international nongovern- mental organizations working on biodiversity conservation (see http://www. conservationmeasures.org/CMP/). Although progress in these areas is promising, in many cases adap- tive management may prove difficult for climate-related decision support, because of the institutional setting of decision making. As shown in Table 3-1, the approach assumes stable goals set by a unitary decision maker that endure for the life of the experiment. Leaders also need to understand the experimental paradigm and to sustain a commitment to experimentation as the learning process unfolds. These conditions are demanding. Deliberation with Analysis The modes of learning discussed above do not take into account two key attributes of the climate-sensitive decision makers needing decision support—that decision environments (climatic and societal conditions and the state of knowledge) change over time and that decisions must consider multiple actors with different objectives and partly conflicting values. De- liberation with analysis addresses these attributes explicitly and, for this reason, we believe it provides the best model for learning in climate-related decision support, though one that still needs further development and research. This model was developed in an earlier National Research Council (1996b) study of decision support in the broad context of environmental risks: Understanding Risk: Informing Decisions in a Democratic Society. The study described a learning process that begins with a broadly based effort involving the range of interested and affected parties to formulate the decision problem (e.g., the risk to be assessed and managed) and to identify the values and interests at stake, so that the likelihood or extent of harm to the system, as well as its various consequences, can be measured or predicted. In this initial deliberation, decision participants interact with

DECISION SUPPORT AND LEARNING 79 technical experts and analysts to develop a shared understanding of the is- sues at stake, of what needs to be understood and how scientific research and assessment and the interpretation of available knowledge are likely to feed into decision making. On the basis of such deliberation, analysts can develop knowledge and information that are likely to be used in further decision-focused deliberations (National Research Council, 1996b). A subsequent National Research Council report (2008c:234) reviewed evidence from multiple sources and concluded that across a broad range of environmental assessment and decision-making contexts, including the 2001 U.S. National Assessment of Climate Change, such “collaborative, broadly based, integrated, and iterative analytic-deliberative processes” provide the method of choice for organizing scientific analysis to serve public decision making. The report also stressed that because of variations in decision contexts, it could not recommend any standard “best practice” way of implementing the model to be applied regardless of the situation. Rather it recommended (National Research Council, 2008c:237): [a] “best process regime consisting of four elements: (1) diagnosis of the context to identify likely difficulties; (2) collaborative choice of techniques to address those difficulties; (3) monitoring of the process to see how well it is working; and (4) iteration, including changes in tools and techniques if needed, to overcome difficulties. In short, it recommended that deliberation with analysis be implemented in different ways in different contexts and structured to learn from its own experience. We, too, conclude that this kind of iterative, analytic-deliberative pro- cess is better suited than other modes of learning to the sorts of chang- ing decision contexts that climate change will present. How well it will work in practice in its various implementations is, of course, an empirical question—one to which techniques of program evaluation could usefully be applied. As the complexity and uncertainty surrounding a risk to be managed increase, so, too, does the degree to which the process of deliberation with analysis becomes iterative. It might proceed through several successive rounds of deliberation by the parties involved and analysis by technical experts on the way to a risk management decision and in reconsidering the decision once implemented (National Research Council, 1996b, 2008c). Each round of analysis and deliberation can yield clearer understanding of the parties’ objectives and of the effects of actions taken. In this sense, the process synthesizes information from assessments of both risks and man- agement actions and thereby informs ongoing decision-making processes. The deliberation with analysis mode of learning resembles program eval- uation and adaptive management, but with two important differences. First,

80 informing decisions in a changing climate the substantive goals of decision support, unlike those of program evaluation, are adopted only provisionally, with an explicit commitment from all par- ties to reconsider them in light of deliberation about what decision support is useful and affordable. The provisional character of the substantive goals resembles adaptive management, although the choice of goals is the result of deliberation rather than a scientific process of hypothesis testing. Second, deliberation with analysis emphasizes wide participation in or- der to generate a consensus among affected parties on what information is needed for decision support, even if not on how to weigh this information. This open process aims to enhance the ability of the organization sponsoring the decision support activity to implement any decisions reached and sustains the ability to continue to deliberate (National Research Council, 2008c). As shown in Table 3-1, the deliberation with analysis approach aims to address situations with a multiplicity of participants, evolving goals, and fluidity in both the natural and social environments. Thus, it presumes a qualitatively different situation than the ones envisioned in conventional program evaluation or adaptive management in which a single decision maker is assumed. As noted above, responding to climate change often in- volves parties with different perspectives, including both the typical differ- ences between scientists and decision makers and the divergent values and interests among the decision makers. As the effects of a changing climate become apparent, the already wide diversity of objectives among affected parties may well increase. Climate change is likely to continue to create the kinds of tensions over decisions and decision support for which deliberation with analysis was developed. Deliberation with analysis has one other possible advantage in the con- text of climate risks. It provides a process that might deal effectively with the great cognitive difficulties of comprehending risks that involve many possible outcomes, over long time periods, and with uncertain probabilities of occurring. Decision analytic tools may help decision makers comprehend such risks, but they may also generate confusion. Similarly, nonspecialists’ modes of cognition may lead to useful simplifications or to confidently held misunderstandings. A process that allows scientists and decision makers to discuss their understandings has the potential to identify and address such problems. Whether or how analytic-deliberative processes can be designed to produce this kind of benefit is a subject for empirical study. Finally, deliberation with analysis acknowledges the need for iterative learning when climate responses and climate-related decision support can both produce surprising outcomes—an approach that has been described by organization theorists as “double-loop” learning (Argyris and Schon, 1978); see Box 3-1. Deliberation with analysis is iterative, and that means that it may not lead quickly to convergence on an answer or action. The learning process

DECISION SUPPORT AND LEARNING 81 BOX 3-1 Double-Loop Learning Effective response to climate change may require transformations in existing management practices, technologies, and organizational structures. The organi- zational analysts Argyris and Schon (1978) pointed out a generation ago that an organization embodies a model of reality—a simplified representation of the situ- ation it faces, embodied in its operating procedures. That model is tested against reality every day. Often, experience produces surprises. Some of those surprises fit with the implicit theory of the organization; for example, when a water management agency faces a drought by calling for conser- vation measures, and then steps up the urgency of its appeals to reduce demand until the rains resume. Making responses that reaffirm the basic model built into an organization is a pattern of adaptation that Argyris and Schon called “single- loop learning.” The organization can learn because the problem is recognized, a solution is implemented, and it works—a feedback loop in action. Sometimes, however, solving a novel problem may require steps outside the organization’s model. A severe drought may prompt proposals to reuse water from treated sewage, for example, engendering conflict that threatens the water agen- cy’s budget or leadership. Finding responses to novel problems is what Argyris and Schon dubbed “double-loop learning”: Adoption of those responses requires an organization to revise not only its practice but also its operating theory, its rules and culture. There are two feedback loops needed in facing such problems: one to overcome existing commitments and one to develop and adopt a response to the problem. Double-loop learning is a change process, often a wrenching one. A changing climate is likely to produce many situations requiring double-loop learning. can sharpen conflicts among participants by clarifying who wins and who loses if particular choices are made. In addition, the analytic process often increases rather than decreases perceived uncertainty, as more precise ques- tions lead to more detailed and elaborate information about what is not known. In these ways, learning can raise the cost of decision making and delay the formation of consensus or consent. The deliberation with analysis model as developed in 1996 strongly em- phasized broad public participation as a way to achieve an actionable un- derstanding of the choices facing a decision-making body—in other words, as a part of decision support. More recently, a set of principles has been identified for effective public participation in environmental decision mak- ing (National Research Council, 2008c); see Box 3-2. Many of these prin-

82 informing decisions in a changing climate BOX 3-2 Lessons for Decision Support from the Study of Public Participation When government agencies engage in public participation, they should do so with • clarity of purpose, • a commitment to use the process to inform their actions, • adequate funding and staff, • appropriate timing in relation to decisions, • a focus on implementation, and • a commitment to self-assessment and learning from experience. Process design should be guided by four principles: 1. inclusiveness of participation, 2. collaborative problem formulation and process design, 3. transparency of the process, and 4. good-faith communication. These elements of design are appropriate to all participatory processes, although the way they are implemented will vary across contexts. There is no single best format or set of procedures for achieving good outcomes in all situations. Decisions with substantial scientific content should be supported with collab- orative, broadly based, integrated, and iterative analytic-deliberative processes (i.e., deliberation with analysis). In designing such processes, the responsible agencies can benefit from following five key principles for effectively melding scientific analysis and public participation: 1. Ensure transparency of decision-relevant information and analysis. 2. Pay explicit attention to both facts and values. 3. Promote explicitness about assumptions and uncertainties. 4. Include independent review of official analyses and/or engage in a process of collaborative inquiry with interested and affected parties. 5. Allow for iteration to reconsider past conclusions on the basis of new information. SOURCE: National Research Council (2008c:2–3). ciples echo those discussed in Chapter 2. The 2008 study emphasizes that difficulties in implementing the principles often arise in specific contexts and that these difficulties have to be addressed in a process suited to the situa- tion. The study recommends (National Research Council 2008c:237):

DECISION SUPPORT AND LEARNING 83 practitioners, working with the responsible agency and the participants, should adopt a best-process regime consisting of four elements: diagnosis of the context to identify likely difficulties; collaborative choice of tech- niques to address those difficulties; monitoring of the process to see how well it is working; and iteration, including changes in tools and techniques if needed to overcome difficulties. Deliberation with analysis has several implications for climate-related decision support. First, the responsible agencies need to take public par- ticipation seriously in their decision support activities, putting in resources and, more importantly, being ready to learn from and to listen to affected parties. The intensity of engagement with the public should be tailored to the level of public attention and anticipated conflict. Agencies need to establish expectations about how they will use public input in ways that are consistent with their legal authorities and responsibilities (National Re- search Council, 2008c). In some situations, they may need to modify their usual procedures to make it possible to use public input. Second, inclusiveness matters. The implications of a changing climate are becoming apparent to constituencies ranging from agriculture to tour- ism to local governments, and their responses are still taking shape. New groups, such as professional societies and public health agencies, are be- coming participants in decision support, joining those, such as water re- source managers, who have long used products from climate forecasters. As more decision makers recognize the need for decision support, they are likely to need novel information in new forms. Agencies need to work with the emerging constituencies and assist other organizations conducting their own climate decisions support. Third, transparency of content matters. Information for climate-related decision support is often derived from models whose workings are often incomplete and nonintuitive; uncertain in terms of the location, time, and magnitude of forecast events; and difficult for nonspecialists to understand. Transparency, particularly to new constituencies, accordingly requires de- liberate two-way communication and interpretation of science. Fourth, the legitimacy of science relies on its transparency, both to peer experts and the public. Independent peer review of scientific content can provide an important measure of credibility and legitimacy by creating a mechanism to counteract bias, correct error, and reveal the range of com- petent scientific judgment. Since formal external review is a costly process, its use is sensibly limited to situations where the content is likely to be both salient and controversial. Part of the learning that is needed includes a better sense of when to use independent reviews or other approaches to identify error, bias, and conflicting judgments.

84 informing decisions in a changing climate Fifth and most challenging, there is a need to learn and to adapt the way that decision support is provided, based on the experience gained through implementation. Public participation shares two characteristics with program evaluation. Both subject the practices of government agen- cies to scrutiny that can be uncomfortable, and both can alter agency staffs’ initial sense of how best to pursue their missions. These frictions are also signs of learning. What needs to be learned, over time, is how to temper internal judgment with the knowledge that comes from taking the public seriously. This is double-loop learning. Meeting the conditions for effective implementation of deliberation with analysis presents significant challenges. These challenges have been successfully met in many contexts, but decision makers will have to learn how to meet them in the new dynamic decision environments climate change will present. Dedicated research on this problem will help practitio- ners develop effective modes of organizational learning for climate-related decision support. Conclusion 4: Climate-related decision making, especially by public agencies, typically involves multiple participants with varied and chang- ing objectives interacting with uncertain and evolving knowledge. The most appropriate mode of learning under such conditions combines deliberation with analysis. This mode is also quite demanding in its needs for leadership and other resources. Recommendation 3: Federal agencies in their own decision support activities and in fostering decision support by others should use the ap- proach of deliberation with analysis when feasible. This is the process most likely to encourage the emergence of good climate-related deci- sions over time. The federal government should also fund research on decision support efforts that combine deliberation with analysis and that use other appropriate learning models, with the aim of improving decision support for a changing climate. FEDERAL ROLES IN FACILITATING LEARNING The federal government can contribute to adaptive learning in response to climate change in three ways: designing its own decision support activi- ties for learning; encouraging nonfederal decision makers to take climate change into account in various ways; and providing support to enable those decision makers to learn more effectively from their own and others’ efforts to respond to climate change. It is important to emphasize that the national response to climate change will be widely distributed and will involve literally millions of deci-

DECISION SUPPORT AND LEARNING 85 sion makers. Thus, it will be important to provide distributed intelligence about the vulnerabilities and opportunities of decision makers and about the potential value of different decision support activities (Lempert, 2007). Doing this is likely to increase the pace of learning and lower its social cost. Distributed intelligence can take many forms. For example, it can in- volve information clearinghouses, monitoring systems, and advisory bodies, organized for long-term consistency and to help translate varying experi- ences into useful guidance for new decisions. Arvai et al. (2006a) suggest that the IPCC holds considerable promise as a reporting body for assessing the wide range of experiments that have occurred around the world if the IPCC and the Secretariat of the U.N. Framework Convention on Climate Change could be strengthened to provide concrete guidance on methods and approaches for adaptation and management. Internet-based mecha- nisms such as blogs, Wikis, and user-based reporting systems may also help provide distributed intelligence on decision support innovations. The rest of this section illustrates some of the strategies the federal government might use to facilitate adaptation and learning by others: sup- porting the diffusion of innovations, using price and quality signals to guide consumer behavior, and supporting networks and boundary organizations; see Table 3-2. Supporting the Diffusion of Innovations Since the nineteenth century, the federal government has provided deci- sion support by fostering the adoption of new technologies and practices, notably in agriculture. An extensive research literature on the diffusion of innovations (Rogers, 2003) offers useful lessons for climate change decision support. The classic example is federal support for the diffusion of agri- cultural innovations through the land-grant university system, established under the Morrill Act of 1862. Diffusion processes usually begin with innovators outside the federal government, and innovations usually spread by example, though sometimes by persuasion. The federal government can facilitate this process in at least three ways: by supporting the development of innovations (e.g., in the ag- ricultural example, by funding research that produces new crop varieties), by encouraging initial adoption (e.g., by having extension agents work with farmers who are willing to try the new seeds), and by helping spread information about successful innovations. This process reflects the panel’s approach for climate change decision support in that it is user oriented, with ultimate choices determined by those who use the innovation on the basis of information about it. The federal government can help generate innovations in decision sup-

86 informing decisions in a changing climate TABLE 3-2  Federal Roles in Promoting Learning Opportunity for Decentralized Principles of Effective Learning Goal Federal Roles Decision Support Diffusion of Adoption of practices Sponsoring invention User oriented   innovation or products, Promoting diffusion Rely on networks of particularly in Supporting internet- communication production based information exchanges Market signals Guide consumption Create markets Affect user choices choices Support and Use marketing channels implement and tools, such as certification and advertising labeling Flexible in the sense Create scarcity that prices adjust through regulatory controls Networks Promote solving of Participate in some Users populate organizational networks networks and drive problems hindering Fund some network cross-sector and responses to functions in multidisciplinary changing climate important problem problem solving areas Flexible and adaptable, though network may be ephemeral port in serving its agencies’ own constituencies and by supporting innova- tive approaches to decision support for constituencies that cannot obtain it without federal support. Federal agencies have done this: One example is the Regional Integrated Sciences and Assessments (RISA) centers supported by the National Oceanic and Atmospheric Administration and through ef- forts to develop new information to meet the needs of specific sectors (see Box 4-1 in Chapter 4). Boundary organizations such as the RISA centers can promote the use of innovative decision support products by “early adopt- ers” and, by linking decision makers to the early adopters, help spread use- ful decision support innovations. Research on diffusion demonstrates that direct contact with peers and peer organizations is critical to diffusion and plays a role complementary to scientific expertise (Rogers, 2003). In addition to directly supporting boundary organizations that help diffuse decision support innovations, federal agencies can help move these innovations into nonfederal networks that can provide a durable institu- tional structure. For example, some engineering and consulting firms that serve the water management sector are now starting to incorporate feder-

DECISION SUPPORT AND LEARNING 87 ally developed climate change information into their portfolio of specialized services. Such private-sector entities will likely rely on technical support and data from federal agencies, similar to the way the National Weather Service provides meteorological data and expertise that mass media and private firms use to develop services and information for users. Federal agencies could also support organizations that produce and maintain Internet-based information exchanges such as blogs, Wikis, and user-based rating and reporting systems for decision support efforts. Such exchanges should be subjected to research that monitors the ways information on the sites is understood and used and the quality control processes used on the sites. Financial Incentives as Stimuli for Learning Changes in markets, especially in the relative prices of fossil fuels and their alternatives, are important for the mitigation of climate change. Fed- eral government policies—such as cap-and-trade or taxation systems for greenhouse gas emissions—can be forms of decision support in that they send signals to consumers about likely future prices, which are likely to influence decisions. Financial incentives, such as tax credits for renewable energy development, similarly send signals to investors and energy produc- ers. Both kinds of signals induce change toward actions with smaller effects on climate than would otherwise be the case. Of course, price signals alone are so limited as a form of communica- tion that they hardly qualify as decision support. Prices do not directly fos- ter understanding of the wider implications of a changing climate, and they provide little information on which behavioral changes are most efficacious. This is notably the case with household consumers, many of whom appear to harbor systematic misconceptions about how best to reduce energy consumption (Kempton et al., 1985; Gardner and Stern, 2008). Prices are therefore most useful for decision support when combined with other policy instruments, including providing more conventional forms of information (e.g., Gardner and Stern, 1996; Stern, 1986). Required Labeling and Certification Regulations can be used to require the provision of some forms of de- cision support. This approach is illustrated by the Energy Star Program of the Environmental Protection Agency, the organic food label requirements of the U.S. Department of Agriculture, and by many voluntary certifica- tion approaches for sustainably caught fish (see Highleyman et al., 2004), sustainably harvested wood (see Conroy, 2007), and “green” buildings (see http://www.usgbc.org/; also see Cole et al., 2005). Certified products may

88 informing decisions in a changing climate then carry a label that assures buyers that the product was produced in compliance with the certification process. Labels ideally condense complex information about the consequences of a choice into a simple signal. Volun- tary certification has been called private regulation (Bartley, 2007) because government does not directly modify the behavior of economic actors. In a globalizing economy, the potential for labeling to influence behavior across national boundaries is noteworthy. Credible labels often require complex auditing and verification systems built on analyses, standards, and practices that can reach from initial pro- duction to retailer (Cole et al., 2005). Information provided by government sources is essential to both voluntary and government-sponsored labeling and certification efforts, as is technical support for advancing assessment methods, such as life-cycle analysis. These are essential elements of deci- sion support. Labeling and certification can meet a user’s need for simple guidance in consumption choices, and are accordingly useful for organizing large-scale social responses, especially by households and other small actors. Certifi- cation processes tend to be multidisciplinary and need to be well enough institutionalized for labeled products to gain significant market share. Rela- tively little is yet known about how to make labeling systems flexible and capable of continuous improvement and learning. There is accordingly a need for research to illuminate and to strengthen this significant form of decision support. Networks and Boundary Organizations As Chapter 2 discusses, networks and boundary organizations can play essential roles in decision support. We focus here on their relevance to learning to improve decision support and on what the federal government can do to make them more effective for this purpose. Network relationships have grown in importance in knowledge- intensive activities, such as public administration and service industries, in which novel problems or opportunities arise that are outside the experience or craft of the professionals in the organization. Networks provide ways to tap experience in other organizations. They are typically user focused and flexible, as well as capable of crossing organizational and disciplinary lines. Responses to climate change can benefit greatly from good networks: Networks can facilitate decision makers’ access to sophisticated knowledge and information drawn from science, engineering, law, and other profes- sions, as well as to each other’s experiences. The federal government can play an important supporting role in facilitating the networks necessary for climate change decision support by helping reduce the costs of communica-

DECISION SUPPORT AND LEARNING 89 tion and coordination. And federal agencies can also benefit from networks for their own decision support activities. A more effectively networked government can respond more fluently to the multifaceted manifestations of changing climate. Interagency teams can work quickly across the mandates of different government departments without waiting for organizational changes in those departments. Federal agencies can also benefit from partic- ipating in and sometimes supporting networks and boundary organizations that reach beyond the federal government to state and local government, the private sector, and civil society. Federal financial support for networks, particularly those serving con- stituencies with limited resources, has several important benefits. As the long-term relationships built through the RISA centers illustrate, networks that link federally supported researchers with users of the knowledge they produce increase the utility of federally sponsored research on climate phe- nomena and facilitate deliberation informed by analysis. Among the kinds of capabilities likely to be cost-effective are support for convening network participants for face-to-face meetings, such as regional conferences, funding a webmaster for a weblog, providing space on an internet server, and pro- viding start-up funds for networks that might be able to develop nonfederal support for their continued activity once members recognize their value. However, the continuation of a network depends on members’ seeing concrete returns on their participation. As with other innovations, some networks will fail to meet this test. Thus, decision support networks should be designed in the expectation that they may be ephemeral. Conclusion 5: An important role of the federal government in climate- related decision support is to facilitate the development and improve- ment of decision support systems by nonfederal entities. Recommendation 4: Federal agencies and other entities that provide de- cision support should monitor changes in science, policy, and climate- related events, including changes outside the United States, that are likely to alter the demand and opportunities for effective decision sup- port. Knowledge of such changes will help them to learn and improve more rapidly. Recommendation 5: Federal agencies should promote learning by sup- porting decision support networks to share lessons and technical ca- pabilities. This may include support for expanding the capacity of boundary organizations and distributed entities for learning, such as internet sites. The federal investment should be selective and guided by the reality that networks operate satisfactorily only when their mem- bers see concrete benefits from participation.

90 informing decisions in a changing climate We conclude by suggesting that the federal government fund studies of social networking, boundary organizations, and other mechanisms that en- able deliberation with analysis on climate-related response options among public- and private-sector organizations; build on models such as the RISA centers to expand the body of practical experience in using networks and boundary organizations to address the issues of climate change; and work with philanthropies and other nongovernmental organizations to develop innovative ways of coordinating networks and supporting boundary orga- nizations to provide distributed mechanisms for learning to provide climate- related decision support.

Next: 4 Information Needs for Decision Support »
Informing Decisions in a Changing Climate Get This Book
×
Buy Paperback | $53.00 Buy Ebook | $42.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Everyone--government agencies, private organizations, and individuals--is facing a changing climate: an environment in which it is no longer prudent to follow routines based on past climatic averages. State and local agencies in particular, as well as the federal government, need to consider what they will have to do differently if the 100-year flood arrives every decade or so, if the protected areas for threatened species are no longer habitable, or if a region can expect more frequent and more severe wildfires, hurricanes, droughts, water shortages, or other extreme environmental events. Both conceptually and practically, people and organizations will have to adjust what may be life-long assumptions to meet the potential consequences of climate change. How and where should bridges be built? What zoning rules may need to be changed? How can targets for reduced carbon emissions be met? These and myriad other questions will need to be answered in the coming years and decades.

Informing Decisions in a Changing Climate examines the growing need for climate-related decision support--that is, organized efforts to produce, disseminate, and facilitate the use of data and information in order to improve the quality and efficacy of climate-related decisions. Drawing on evidence from past efforts to organize science for improved decision making, it develops guidance for government agencies and other institutions that will provide or use information for coping with climate change. This volume provides critical analysis of interest to agencies at every level, as well as private organizations that will have to cope with the world's changing climate.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!