The ever more apparent impact of human actions on biodiversity in recent decades has led to an international scientific and political effort to measure, preserve and manage biodiversity (e.g. IPBES, EU’s Horizon 2020 program). To achieve this objective, we need a better understanding of the ecological processes that determine natural communities and we need appropriate modelling tools to make accurate predictions about the distribution and dynamic of biodiversity in the face of anthropogenic pressures.
To do so, scientists currently favour correlative statistical methods (e.g. direct relating of species abundance along climatic gradients) instead of process-based approaches that take explicitly into account how climatic gradients influence species demography and interactions and how the latter determine species abundances. The reason for this preference is not so much the conceptual superiority of correlative methods but rather because they couple easily with the available biodiversity data while process-based approaches, which are in principle better suited to estimate local diversity and biodiversity dynamics are currently more challenging to link to the available biodiversity data. Indeed, to calibrate process-based models, we need to measure experimentally a large quantity of demographic features and biotic interactions among the target species. As this is both complicated and costly, this approach has often been limited to simplified experimental settings, thus strongly limiting their use for ambitious biodiversity modelling projects.
This project (CLIMB) proposes a way forward to overcome the limitations of process-based approaches: it proposes that the demographic features and biotic interactions species could be approximated by functional traits. Functional traits are quantifiable traits of organisms that are easily measurable (for instance plant height or plant leaf area) and they have been shown to be directly or indirectly linked to species demography and biotic interactions.
More precisely, CLIMB aims to develop and test a general statistical procedure that will infer the demographic features of species from known functional trait data to best predict the observed structure of biodiversity. This approach will shed light on some of the oldest questions of ecology, but also explore a new route to connect process-based biodiversity models with functional trait data and thus provide much-needed tools for answering pressing issues in biodiversity policy.
