Center for Macroecology, Evolution and Climate

Standards for models in biodiversity assessments

The purpose of establishing standards for models in biodiversity assessments is to provide a hierarchy of reliability, ensure transparency in the translation of science results into policy, and to encourage further improvements in the underlying science. Real-world implementation of the standards requires that detailed and flexible criteria be identified and agreed.

We propose an extensive set of criteria that enable scoring modelling studies available for biodiversity assessments. We define four levels of standards for scoring models in biodiversity assessments.

The gold standard is aspirational. It is difficult or even impossible to achieve at present, usually requiring next-generation data and modelling approaches, and often including manipulative experiments.
The silver standard refers to current cutting-edge approaches, typically involving data or analyses that allow uncertainty (from both data and models) to be reduced or at least estimated.
The bronze standard encompasses data and procedures that represent the minimum currently acceptable practices in the field. It includes established approaches and attempts to approximately characterise and address limitations of data and models.

The deficient category involves the use of unacceptable data and modelling practices. Although models characterised as ‘deficient’ can potentially be useful for addressing specific research questions, the outcomes of these studies should be interpreted with extreme caution within biodiversity assessments.