Decision analysis is a systematic approach to uncertainties in decision-making, using models to project outcomes of alternative actions. These outcomes (or “performance measures”) act as criteria for comparing and ranking the actions. Many possible outcomes are generated for each action, each resulting from a unique combination of alternative hypotheses about uncertain states of nature (e.g., historical data usually suggest a range of ecological responses to management actions; there is always a range of future possible climate regimes). Such an approach has several benefits over approaches in which uncertainties are ignored or treated in an ad hoc manner (e.g., using point estimates for uncertain parameters).
First, this approach can lead to more risk-averse management decisions in the long run because it allows decision-makers to select “robust” actions that perform well under a broad range of assumptions. Implementing actions that are robust to uncertainties minimizes risks that the selected action will not have its intended effect, because such actions have a much higher chance of producing a favourable outcome even if the underlying hypotheses turn out to be wrong.
Second, including uncertainties allows risk-averse actions to be identified and implemented before these uncertainties are fully resolved. This avoids the common situation where uncertainty in the predicted effects of management actions discourages decision-makers from changing management policies from the status quo (Peterman and Anderson 1999). Avoiding such inaction is particularly important for situations in which management actions must be taken long before uncertainties can be resolved to the satisfaction of decision-makers.
Third, explicitly incorporating alternative hypotheses about key uncertainties supports collaborative processes involving scientists and agencies with alternative interpretations of existing data.
ESSA’s Experience in Applying Decision Analysis to Resource Management
ESSA has used decision analysis procedures since 1995 to facilitate stakeholder agreement on management actions, to help scientists agree on which uncertainties most critically affect decisions, and to assess the benefits and costs of adaptive management actions.
Decision analysis is a powerful approach to making decisions in the face of significant uncertainty. BC Hydro has recognised this by incorporating a decision analysis approach into the Water Use Planning Process. ESSA has successfully applied decision analysis to a wide range of multi-stakeholder decision-making processes in many areas of resource management. While a powerful tool in its own right, we have found that the utility of decision analysis is greatly enhanced when used in conjunction with other skills at which ESSA excels, including multi-stakeholder facilitation, experimental design, modelling, and technical analyses.
For example, ESSA was the lead facilitator for a multi-agency research program (PATH) to evaluate the effects of proposed operation of the Federal Columbia River Hydropower System on ESA-listed salmon species in the Snake and Columbia River (Marmorek and Peters 2001, Peters and Marmorek 2001, Peters et al. 2001). As facilitators, we used decision analysis as a tool for involving scientists and agencies with varying interests, values, and viewpoints by allowing multiple hypotheses about the role of hydrosystem, climate, habitat, and other effects on observed declines of Snake River salmon (Peters et al. in review; Peters and Marmorek in review; Deriso et al. in review). Using this approach, we were able to make progress on identifying key uncertainties and the effects of actions despite the long-standing issues and uncertainties that had stymied previous co-operative efforts in the high-stakes arena of Columbia River salmon management.
We recently completed a similar decision analysis / facilitation approach to assist BC Hydro and other stakeholders in the development of a Water Use Plan for the Cheakamus River. This WUP process has been challenging due to several factors including the “anchoring” created by a previous court case, and difficulties faced by the Fisheries Technical Committee in identifying appropriate fish population performance measures. Our facilitation expertise has maintained a high level of commitment to the WUP process by all members of a stakeholder Consultative Committee over a two and a half year period, despite historic tensions. Also, ESSA’s experience in leading impact hypothesis workshops helped the FTC to prioritise research needs, develop models and performance measures, and provide evaluations of alternatives that helped the Consultative Committee evaluate alternatives (Marmorek and Parnell 2000).
In another project for BC Hydro, we used decision analysis to evaluate effects of alternative releases from Keenleyside Dam on dewatering mortality of mountain whitefish populations (Alexander et al. 2000a). The decision analysis was embedded in a user-friendly software tool that could be used not only to evaluate proposed long-term release strategies, but also to assess potential experimental releases from Keenleyside Dam and the trade-offs among the potential learning benefits of those experiments, their costs in terms of foregone power revenues, and the risks imposed on whitefish populations. The decision analysis software we developed served as a useful vehicle for incorporating the views and values of scientists from BC Hydro, MELP, DFO, and First Nations.
As a final example, ESSA is using decision analysis to assess experimental releases from Whiskeytown Dam (Clear Creek, California) on downstream ESA-listed chinook and steelhead populations (Alexander et al. 2000b). One of the challenges in this project is to consider trade-offs not only between fish and power-related values, but also among operation of various projects in the same river system. For example, increasing flows from Whiskeytown Dam may require decreasing releases from other dams in the Sacramento power system, which will affect power generation and downstream fish populations associated with those dams. Decision analysis is providing a useful structure for considering these types of complex interactions between operations at connected hydroelectric projects.
These examples provide a broad overview of ESSA’s experience in using decision analysis as a method for evaluating actions. A key element in the success of these projects has been ESSA’s ability to combine decision analysis with our expertise in the area of facilitation and mediation, statistical and technical analyses, and modelling. Taken together, this suite of tools, skills, and experience provides a powerful and effective approach to evaluating trade-offs and making resource management decisions in a multi-stakeholder environment.
- Decision analysis of actions affecting endangered Snake River chinook salmon.
- Use of decision analysis to design adaptive flow management experiments in Clear Creek, northern California; US Department of Interior, Bureau of Reclamation.
- Decision analysis of adaptive management experiments for Columbia River whitefish management; BC Hydro.
- Decision analysis in a multi-stakeholder water use planning process for an hydroelectric facility; BC Hydro.
Alexander, C.A.D., C.N. Peters, D.R. Marmorek and P. Higgins. 2006. A decision analysis of flow management experiments for Columbia River mountain whitefish (Prosopium williamsoni ) management. Can. J. Fish Aquat. Sci. 63: 1142–1156.
[PDF - 658 kb]
Marmorek, D.R. and C.N. Peters. 2001. Finding a PATH towards scientific collaboration: insights from the Columbia River Basin. Conservation Ecology 5(2): 8.
[HTM - off site]
Peters, C.N. and D.R. Marmorek. 2001. Application of decision analysis to evaluate recovery actions for threatened Snake River spring and summer chinook salmon (Oncorhynchus tshawytscha). Can. J. Fish. Aquat. Sci. 58(12): 2431-2446.
[PDF- 2.02 mb]
Peters, C.N., D.R. Marmorek, and R.B Deriso. 2001. Application of decision analysis to evaluate recovery actions for threatened Snake River fall chinook salmon (Oncorhynchus tshawytscha). Can. J. Fish. Aquat. Sci. 58(12): 2447-2458.
[PDF- 1.57 mb]
Peterman, R.M. and C.N. Peters. 1998. Decision analysis: taking uncertainties into account in forest resource management . In: Sit, V. and B. Taylor (eds.). 1998. Statistical Methods for Adaptive Management Studies. Res. Br., B.C. Min. For., Victoria, B.C. Land Manage. Handb. No. 42.
- For more information please contact: David Marmorek