Technology transfer has often been approached as a marketing problem. Researchers view the technology they have developed as a product that they must "sell" to potential users, who are often referred to as "customers" or "clients." This paradigm sets up a buyer-seller mentality that can hinder the successful adoption of complex technology by managers. We believe that the transfer of complex decision-support models presents at least seven formidable difficulties:
• Teaching managers or their support staff to run modeling software requires formal training and technical support. This is not unlike the process of learning any new software, but in addition, it requires an explanation of basic modeling concepts.
• Proper application of models by managers requires that they understand in some detail the assumptions behind the models and the limitations of the results. There is no small danger that inappropriate conclusions can be drawn should users of the models misunderstand the key underlying assumptions.
• Managers must learn how to interpret a model's results to provide defensible support for their decisions.
• Managers must precisely explain to researchers the decisions they must make and the information or knowledge required to make those decisions. Such an understanding will help researchers to judge whether the model can in fact produce the information that is needed.
• Political, funding, or logistical limitations may constrain management options. These issues may not be apparent to researchers, so managers must identify them; researchers could also interview the managers to identify any relevant constraints.
• Researchers may be unfamiliar with the specific land base that is being managed, and may therefore be poorly equipped to accurately model the ecological or management dynamics.
• Managers and researchers may have different understandings of uncertainty and of the risks associated with that uncertainty. A shared understanding of the role of uncertainty in the decisionmaking process is critical.
A common thread among these difficulties is the necessity for substantive communication and partnership between researchers and managers.
To resolve these difficulties, we adopted a collaborative, iterative approach for technology transfer (Ahern 2002, Fall et al. 2001). The approach is collaborative because it assembles a triad of researchers, management planners, and local resource experts. It is iterative because communication among the triad partners must occur repeatedly, so that the application of the tool can be refined with each iteration. Our approach fosters a "community of practice" in which people build understanding together in a social, physical, and temporal setting (Allee 1997, p. 219).
The conceptual framework for our collaborative, iterative approach to technology transfer is best represented as a triangular interaction among researchers, management planners and decisionmakers, and local resource experts (Figure 3.1). The interaction takes the form of iterative communication (the arrows in Figure 3.1), and the focus of the communication is on application of the modeling technology to support a particular management decision. Thus, the modeling tools serve as a common framework that lets all parties conceptualize and formalize (i.e., model) the management problem. The end result is a more defensible decision and a transfer of modeling technology from a research environment to a management environment. This approach also fosters the development of a shared vision of the model's requirements and the decision process to be supported, of the data requirements, of the interactions among ecological and human processes, of the model's capabilities and assumptions, and of how to appropriately interpret the model's outputs.
The collaborative nature of the process is important because each partner provides expertise that is critical to successful technology transfer. The researchers understand the feasibilities of applying existing models or building new models, know the assumptions that underlie a model, are familiar with the algorithms that drive a model, know how to estimate the model's parameters and develop the input data, provide the technical expertise to run a model, and guide interpretation of the
model's results. Models are not a mysterious black box to the researchers who developed them, and this knowledge helps them to make the models more transparent to the other partners, giving them greater confidence in the output. The management planners understand the management decision that must be made, and can readily identify the information gaps that hamper their ability to make defensible decisions. They also can identify the bounds of politically or logistically feasible alternatives. Without such input, researchers are likely to develop elegant and sophisticated answers to irrelevant questions. The local resource experts enable the model application to reflect the best current knowledge about the system under study. They help the researchers to estimate realistic values of model parameters for the local ecosystems. They can readily identify model behaviors that incorrectly simulate the local reality and assist the planners to develop ecologically feasible management options. A two-way collaboration that only includes the researchers and planners is more likely to result in biologically indefensible results.
The collaborative interaction begins with a meeting of all three groups. The initial iteration focuses on sharing of information about the management decision to be made, the alternatives that will be compared, data availability, and the modeling tool to be applied. Resource experts inject biological reality into the discussion. Following this meeting, the researchers design the modeling protocols, work with the resource experts to estimate the model parameters, oversee the generation of input data, and perform the initial simulation runs. During this time, the researchers contact resource experts to clarify and refine the initial parameters. The second iteration brings all parties together again to review the initial model outputs. The resource experts assess whether the model's behavior is consistent with their understanding of the ecological system. The planners assess whether the information generated by the model is what they need to make a decision, and if not, work with the researchers to refine the modeling objectives. The researchers communicate any needs for better parameter estimates or additional information about management alternatives. A second round of simulations is then conducted based on the improved understanding of the management problem. The process is repeated until all parties are satisfied that the model is producing the information required to make the decision. Collaboration and iteration produce several important outcomes:
• The model results are of greater quality and relevance to the decisionmaker.
• Managers learn to use a new technology.
• Researchers learn about management problems and the constraints that managers face.
• Resource experts come to better appreciate the interactions among many resources and the realities of multiple-use planning.
The collaborative nature of the approach provides the synergy required for effective technology transfer.
To illustrate the application and utility of the collaborative, iterative approach to technology transfer, we will describe two case studies in which modeling technologies were successfully transferred from a research and development environment to operational use, to meet a management decision-support need. The second case study is a technology-transfer effort in progress. The first used the SELES modeling language to construct a new model that would support a complex and controversial land-use planning process. The second used the LANDIS model to predict how patterns of human settlement and forest management in the Nicolet National Forest (Wisconsin, USA) might intersect to influence fire risk. Before presenting the case studies, we have provided a brief orientation to the modeling technologies used and the philosophy behind their development to describe the basic principles of the underlying science.
Was this article helpful?