Species distribution models (SDMs), which relate species occurrence records to environmental datasets, are increasingly advocated in the peer-reviewed literature to support conservation decisions. Nearly every paper that uses SDMs has a token section about how the developed SDMs could be used in future conservation efforts. But is there much evidence that SDMs are actually being used to support conservation decisions on the ground? How could we most effectively develop SDMs to support these decisions, and start to bridge the gap between academics and managers?
In a recent paper in Ecology Letters, several authors from the ARC Centre of Excellence for Environmental Decisions (including Tracey Regan, Brendan Wintle and John Baumgartner from QAECO) take the first steps toward tackling these challenges. In our manuscript, we use examples from four important conservation domains to illustrate how the development of SDMs should be dictated by the decision context in which they are to be used, and describe some of the types of constraints that need to be considered in different decision contexts. Drawing on the unpublished conservation literature and our own experiences in the realm of conservation decision making, we show that SDMs are being used to support conservation efforts, but suggest that SDMs could be used more effectively by developing them within a structured decision making framework that considers the real world constraints faced by managers.
So if you’re doing research on SDMs – get out there and get involved in a real conservation decision! Your research will help inform a practical conservation problem, and you may just find that the experience provides exciting new collaborations and directions for your own research.
You can download our paper free of charge here.
Relative likelihood of occurrence of an Australian frog according to MaxEnt, a popular type of SDM. How one builds such a model, and how it will be used to inform a conservation decision (e.g. critical habitat designation) should depend upon the decision context.