Qualitative value of information provides a transparent and repeatable method for identifying critical uncertainty

Abstract

AbstractConservation decisions are often made in the face of uncertainty because the urgency to act can preclude delaying management while uncertainty is resolved. In this context, adaptive management is attractive, allowing simultaneous management and learning. An adaptive program design requires the identification of critical uncertainties that impede the choice of management action. Quantitative evaluation of critical uncertainty, using the expected value of information, may require more resources than are available in the early stages of conservation planning. Here, we demonstrate the use of a qualitative index to the value of information (QVoI) to prioritize which sources of uncertainty to reduce regarding the use of prescribed fire to benefit Eastern Black Rails (Laterallus jamaicensis jamaicensis), Yellow Rails (Coterminous noveboracensis), and Mottled Ducks (Anas fulvigula; hereafter, focal species) in high marshes of the U.S. Gulf of Mexico. Prescribed fire has been used as a management tool in Gulf of Mexico high marshes throughout the last 30+ years; however, effects of periodic burning on the focal species and the optimal conditions for burning marshes to improve habitat remain unknown. We followed a structured decision‐making framework to develop conceptual models, which we then used to identify sources of uncertainty and articulate alternative hypotheses about prescribed fire in high marshes. We used QVoI to evaluate the sources of uncertainty based on their Magnitude, Relevance for decision‐making, and Reducibility. We found that hypotheses related to the optimal fire return interval and season were the highest priorities for study, whereas hypotheses related to predation rates and interactions among management techniques were lowest. These results suggest that learning about the optimal fire frequency and season to benefit the focal species might produce the greatest management benefit. In this case study, we demonstrate that QVoI can help managers decide where to apply limited resources to learn which specific actions will result in a higher likelihood of achieving the desired management objectives. Further, we summarize the strengths and limitations of QVoI and outline recommendations for its future use for prioritizing research to reduce uncertainty about system dynamics and the effects of management actions.