Procrustean beds and empty boxes: on the magic of creating environmental data

This article explores a dark side of the current enthusiasm for compiling large datasets in support of evidence-based conservation, at various scales. We use a series of concrete examples to show how data gathered in biodiversity databases can be poorly informative for the design and implementation of effective conservation strategies and actions. This is due to two mechanisms. The first is the use of ill-advised formats, compelling contributors to fill in data which are at odds with the ecological information that they are truly able to capture. The second corresponds to the fact that, unless knowledge gaps are explicitly and emphatically highlighted, the elaboration of databases bestows visibility on knowledge and invisibility on knowledge gaps. Given these risks, we call for a cultural shift among conservation practitioners, consultants and others, to embrace the idea that documenting and acknowledging knowledge gaps and uncertainties is just as important as compiling data and taking known information into account. We also propose a series of concrete reforms to address these risks. In particular, we point the need for procedural improvements in the process through which conservation databases are elaborated and we suggest introducing differentiated data sharing policies for conservation databases