A bivariate model for estimating reciprocal causality within biogeomorphic ecosystems: The cross‐lagged panel model

This article also appears in: Biogeomorphology, quo vadis? On Processes, time, and space in biogeomorphologyInternational audienceBiogeomorphic ecosystems (e.g. rivers, salt marshes, mangroves and coastal dunes) are shaped by feedbacks between geomorphology and engineer plants that occur at various spatiotemporal scales. The classical bivariate and multivariate statistical methods currently used in biogeomorphology do not permit clear identification of reciprocal causality between geomorphic and biological variables. The aim of this article is to present the potential of the cross-lagged panel model (CLPM) to estimate reciprocal associations (causality) between one geomorphic and one biological variable over time. This tool, which originates from behavioural, social, medical and educational sciences, has clear potential as a novel approach to causal analysis in the context of biogeomorphic ecosystems. We provide a case study of the application of CLPM for analysing biogeomorphic feedbacks between topography and Populus nigra L. physiognomy on a wooded point bar of the Garonne River, France