sustainable economic vitality, and increasing fundamental scientific understanding.
Each piece of infrastructure enables or supports a set of data collection and/or modeling activities, and therefore supports the production of information, which has value. The same piece of infrastructure also has a cost associated with it (e.g., building and maintaining a ship or computer model, training and supporting a technician, archiving and making accessible a data set). The task of prioritizing ocean research infrastructure investments can be interpreted as maximizing net benefits over time by choosing the best combination of infrastructure investments needed to address the science within budget constraints. The committee concedes that there may be other legitimate considerations beyond those spelled out in this report, but most likely these could all be incorporated into an economic optimization framework.
The bottom-up linkage from infrastructure to societal benefits shown in Figure 1.1 provides a useful approach to thinking about infrastructure priorities. An important feature of this prioritization is economy of scale and scope, as a given piece of infrastructure may support multiple research activities, models, and science questions. For example, a particular mooring may support multiple sensors, each sensor can supply data that feed into several models, and each model can contribute information to one or more societal objectives. In addition, a system of coordinated sensors can provide information that is more valuable than their individual contributions. An approach of this kind requires knowledge about the value (benefit) generated by specific information about the ocean and its contribution to achieving societal objectives; linkages between each piece of infrastructure and this specific information; and the cost of each piece of infrastructure.
The value of information (Howard, 1966; McCall, 1982; Nordhaus, 1986) relevant to societal objectives is determined by the degree to which the information allows decision makers to achieve an economically better outcome. The role of information is to reduce the uncertainty under which these decisions are made. For example, the societal objective of managing the nation’s commercial marine fish stocks for maximum sustainable yield can be advanced by improving the quality of information represented by stock assessments and forecasts of fish stock abundance under different levels of fishing effort, environmental conditions, and ecological interactions. When information (e.g., stock assessments, forecasts, interactions between species within and across tropic levels) is less than perfect, fisheries managers must make decisions with greater uncertainty. Uncertainty can be addressed by either reducing the fish catch below the theoretical sustainable yield or by accepting an increased risk that the stock will be overexploited. If these assessments and forecasts were perfect, fisheries managers could allow fishing closer to maximum sustainable yield without risking overexploitation or other adverse ecological consequences. By increasing yield without reducing sustainability, the economic value of the fish stock to the nation could be maximized. The difference in economic outcomes with and without the information is its value.
Although infrastructure costs can usually be determined with considerable accuracy, the value of information in most cases can only be estimated (e.g., Adams et al., 1995; Nordhaus and Popp, 1997; Teisberg and Weiher, 2000; Williamson et al., 2002). Certainty about the value of information from research investments decreases the further it is removed from helping to answer specific applied questions; this uncertainty is greatest for basic science investments, where the nature of the answers and their applications are by definition not well identified in advance. Uncertainty about the expected value of information from infrastructure investments arises from several sources, including uncertainty about the performance of new technologies, the nature of information generated by new technologies or research activities, and the value that the information will in fact generate. Uncertainty can lead to missed opportunities in commercial market assessments, when comparing a well-known market with an arguably better but less well defined market (e.g., the Innovator’s Dilemma [Christensen, 1997]). Deep-mapping autonomous underwater vehicles (AUVs) provide an ocean technology market example. They are an example of a disruptive new technology introduced to the established seafloor survey market, which had relied upon deep towed systems prior to AUV use. Due to the established companies’ hesitancy in adopting a new technology or because of their already significant investment in the existing technology, smaller survey companies using AUVs were able to quickly gain a strong market.
It is not necessary to have perfectly accurate estimates of the value of information in order to make reasonable prioritization decisions. It is necessary, however, to employ a rigorous and harmonized approach that will need to be undertaken at a national level—one that is consistent across and between all relevant agencies, and one that treats uncertainty about returns from investments in a systematic way. Uncertainty in making ocean research infrastructure choices can be addressed in part through mechanisms for the treatment of uncertainty in investment decisions (Dixit and Pindyck, 2010), and the emerging theory and practice of strategic decision making about real options in research and development (Trigeorgis, 1996). Much of this work is focused on investment in research and development by firms seeking to maximize profits from future technology improvements (Bowman and Moskowitz, 2001; Huchzermeier and Loch, 2001; Weeds, 2002; Gunther-McGrath and Nerkar, 2004; Wang and Hwang, 2007), but these problems are structurally analogous to the challenge facing government agencies as they seek to maximize return from research infrastructure investments.