Towards causal relationships for modelling species distribution
(2023) Journal of Biogeography — Vol. 51, n° 5, p. 840-852 (2023)
Files
JournalofBiogeography-2023-DaRe-Towardscausalrelationshipsformodellingspeciesdistribution.pdf
Open Access - Adobe PDF
- 12.32 MB
Details
- Authors
Da Re, Daniele 
Tordoni, Enrico Lenoir, Jonathan Cornejo Rubin De Celis, Sergio Steven
- Abstract
- Aim: Understanding the processes underlying the distribution of species throughspace and time is fundamental in several research fields spanning from ecology tospatial epidemiology. Correlative species distribution models rely on the niche con-cept to infer or explain the distribution of species, though often focusing only on theabiotic component of the niche (e.g. temperature, precipitation), without clear causallinks to the biology of the species under investigation. This might result in an over-simplification of the complex niche hypervolume, resulting in a single model formulawhose estimates and predictions lack ecological realism.Location: Not applicable.Time Period: Not applicable.Major Taxa Studied: Virtual species.Materials and Methods: We believe that a causal perspective associated with a finerdefinition of the modelling target is necessary to develop more ecologically realisticoutputs. Here, we propose to infer the geographical distribution of a species by ap-plying the modelling relation approach, a causal conceptual framework developed bythe theoretical biologist Robert Rosen, which can be formalized through structuralequation modelling.Results: Our findings suggest that building a model relying on a strong conceptualbasis improves the stability of the estimated model's coefficients, without necessarilyincreasing the predictive accuracy metrics of the model.Main Conclusions: Including causal processes underlying the spatial distribution ofspecies into an inferential formal system highlights the methodological steps whereuncertainty can arise and results in model outputs which are tightly linked to the ecol-ogy of the target species.K E Y W O R D Sdirected acyclic graph, environmental niche models, habitat suitability models, path analyses,process-based models, Robert Rosen, statistical models, virtual species
- Affiliations
Citations
Da Re, D., Tordoni, E., Lenoir, J., Cornejo Rubin De Celis, S. S., & Vanwambeke, S. (2023). Towards causal relationships for modelling species distribution. Journal of Biogeography, 51(5), 840-852. https://doi.org/10.1111/jbi.14775 (Original work published 2023)