Creating usable science: Opportunities and constraints for climate knowledge use and their implications for science policy

Abstract

In the past several decades, decision makers in the United States have increasingly called upon publicly funded science to provide “usable” information for policy making, whether in the case of acid rain, famine prevention or climate change policy. As demands for usability become more prevalent for publicly accountable scientific programs, there is a need to better understand opportunities and constraints to science use in order to inform policy design and implementation. Motivated by recent critique of the decision support function of the US Global Change Research Program, this paper seeks to address this issue by specifically examining the production and use of climate science. It reviews empirical evidence from the rich scholarship focused on climate science use, particularly seasonal climate forecasts, to identify factors that constrain or foster usability. It finds, first, that climate science usability is a function both of the context of potential use and of the process of scientific knowledge production itself. Second, nearly every case of successful use of climate knowledge involved some kind of iteration between knowledge producers and users. The paper argues that, rather than an automatic outcome of the call for the production of usable science, iterativity is the result of the action of specific actors and organizations who ‘own’ the task of building the conditions and mechanisms fostering its creation. Several different types of institutional arrangements can accomplish this task, depending on the needs and resources available. While not all of the factors that enhance usability of science for decision making are within the realm of the scientific enterprise itself, many do offer opportunities for improvement. Science policy mechanisms such as the level of flexibility afforded to research projects and the metrics used to evaluate the outcomes of research investment can be critical to providing the necessary foundation for iterativity and production of usable science to occur.

Introduction

As the problem of climate change rises in public policy agendas around the world, the need for robust science to inform policy design also increases. And while the production of climate science has steadily grown (NRC, 2007, p. 94; IPCC, 2007), in the United States, its usability remains relatively limited in terms of decision support and policy design (NRC, 2009a, NRC, 2009b). For example, in 1990, the United States Congress established in law the U.S. Global Change Research Program (USGCRP) and called for the provision of “usable information on which to base policy decisions relating to global change” (US Congress, 1990). The law referred to usable information as knowledge that could be “readily usable by policymakers attempting to formulate effective strategies for preventing, mitigating, and adapting to the effects of global change” (Ibid.). However in both provisions, the US Congress did not specify what it meant by usable, nor did the law suggest how the program should evaluate its effectiveness in terms of usability (Pielke, 1995).

Funded by the federal government through 13 different agencies,¹ through the years, the USGCRP² has mostly focused on fundamental science. Until 2009, the Program's implementation was organized around seven main scientific priorities (climate dynamics, ecosystems, atmospheric composition, the water cycle, the carbon cycle, land use change, and human dimensions) and operationalized by seven interagency working groups (one for each element).³ A committee—chaired by a high level member of one of the participating agencies and under the auspices of the Executive Office of the President (EOP)—guides the program and a small integration office coordinates its activities. The budget for the USGCRP is allocated to each agency independently, although there is some effort to coordinate activities through the EOP, the interagency working groups, and the Coordination and Integration office. Although the level of emphasis on the scientific themes has changed over time (for example to expand the scope of human dimensions of climate change or to boost research focusing on decision-making tools), research focusing on physical and environmental aspects of climate has overwhelmingly dominated the program's budget (NRC, 2009b, Dilling, 2007). And with the exception of the National Climate Assessment completed in 2000, the input of potential users of the information in shaping the USGCRP research agenda has been limited (NRC, 2007).

While the scientific accomplishments of the USGCRP in understanding the climate system have been amply acknowledged (NRC, 2007, NRC, 2010), in recent years, the Program has come under greater scrutiny both from those who first created it (the U.S. Congress) and from scholars who have analyzed different aspects of its design and implementation (Lambright, 1997, NRC, 2009b, NRC, 2010, Pielke, 1995). In 1992 and 2002, Congress critiqued the USGCRP for being “less than successful at developing information that is useful to policy-makers and resource managers in making informed decisions” (Pielke, 1995, HCS, 2002, p. 5). Five years later, a review from the National Research Council of the US National Academy of Sciences (NRC/NAS), found that “inadequate progress has been made in synthesizing research results, assessing impacts on human systems, or “providing knowledge to support decision making and risk analysis” (emphasis added, NRC, 2007, p. 34). In 2009, two other NRC/NAS reviews have reiterated that the program has fallen short from the goal of supporting policy and decisions on the ground (NRC, 2009a, NRC, 2009b).

Central to this critique is how one evaluates the usability of science, that is, what it means for science to be usable in the context of decision-making. While basic science may become applied science in the future, or eventually support whole new industries or technologies that we cannot even imagine today (Stokes, 1997), we argue that its function differs from that implied by the call for usable science for decision making. Alternatively, usable science is that produced to contribute directly to the design of policy or the solution of a problem (Lemos and Morehouse, 2005, NRC, 2009a, NRC, 2009b, Weiss, 1978). This implies a much more specific, time sensitive role for science to be used in supporting decisions as they exist today or in the near future, and appears consistent with the US Congress’ intent that the USGCRP provide information “readily usable by policymakers attempting to formulate effective strategies.” In this context, it is not about which science is more important or about usable science replacing basic science—both are necessary and many times complement each other (Lemos and Morehouse, 2005, NRC, 2010). Rather, we focus particularly on usable science for decision making as stated by the Global Change Research Act, and the USGCRP (whose mission includes research, education, communication and decision support [emphasis added]).⁴

In this paper, we suggest that usability exists within a range in which each use is defined by a perception of usefulness and the actual capacity (e.g. human and financial resources, institutional and organization support, political opportunity) to use different kinds of information. In a recent paper, Lemos and Rood (2010, p. 673) describe this range by arguing that

(…) different actors perceive the usefulness of scientific information differently. Scientists, for example, when choosing the focus of their research, may make an assumption of what they think decision makers need and hope their work will meet that need. Users, in turn, may define their need differently. However, scientists and users do not uniformly make the same assumption about what they think is useful and what they know is usable. Thus, some scientists’ assumptions may be closer to users’ definition of need, while others’ may be farther away. In this sense, there is a range of perceptions of usefulness and usability.

Providing information that is “readily usable” for decision making must therefore navigate and bridge any differences that might exist between what scientists might think is useful, and what is actually usable in practice. This entails establishing a shared vision of what knowledge is usable in a given decision process. We can think of the production and uptake of scientific knowledge as a pull–push process in which different conditions, mechanisms and institutions shape ultimate usability. Here we argue that usability is a function of both how science is produced (the push side) and how it is needed (the pull side) in different decision contexts. We further suggest usability is most effectively pursued through deliberate science policy design and implementation, or “reconciling supply and demand,” as needs for information are often not well met through the independent production of scientific knowledge alone (Sarewitz and Pielke, 2007). One critical aspect of this design is the creation of the conditions and mechanisms that enable iterativity, that is, the purposeful and strategic interaction between climate knowledge producers and users so as to increase knowledge usability (Lemos and Morehouse, 2005). In this article, we contend that the creation of these conditions and mechanisms is predicated on the action of organizations and actors that take upon themselves the responsibility to build them. In other words, these actors and organizations ‘own’ the task of fostering iterativity rather than expecting it to fall on someone else's shoulders or to be a consequential outcome of the call for the creation of usable science alone. And whereas in some cases factors constraining or fostering use of climate information may be outside the purview of what the scientific enterprise can influence and control (e.g. organizational constraints, lack of human resources, lack of political support), in other cases, they might be within the scope of what science policy can effect. Hence, to produce usable science, effective science policy should foster iterativity not only by purposefully incentivizing producers and users to own the task of creating it but also by eliminating the constraints that inhibit it.

In the next few sections, we explore opportunities and constraints for science use both in the way science is produced as well as in the way it is needed and used. We rely on evidence from different areas of climate information use but especially seasonal climate forecasting (SCF), about which there is a rich empirical literature. We aim at exploring evidence-based assumptions of different factors that influence the likelihood of climate science being used in decision making and policy. First, we synthesize ‘lessons learned’ from the literature—both from a more general perspective and from the perspective of climate science—seeking to understand the factors that shape information use; second, we explore how these lessons can inform climate science policy. In section two, we briefly discuss different modes of science production and illustrate the models that inform this analysis. Section three summarizes the factors constraining and fostering climate science use based on the empirical literature. In section four, we speculate how lessons learned from empirical cases can inform the design of science policy to foster the usability of climate science for decision-making.