A graphical causal model for resolving species identity effects and biodiversity–ecosystem function correlations
Open Access
Abstract
Identifying and clearly communicating the drivers of ecosystem function is a crucially important goal for both basic and applied ecology. This has proven difficult because the putative causes (e.g., environment, species identity, biodiversity, and functional traits) are numerous and correlated. The problem is exacerbated by a lack of a formal framework for unambiguously relating theoretical language to precise, quantitative expressions of that language. Using a formal framework for the graphical expression of complex causal hypotheses, we developed a causal diagram of the concepts required to comprehensively test whether hypothesized sets of functional traits mediate the relationship between community structure and ecosystem function. We then used causal analysis, simulations, and field data to develop and test analytical strategies for understanding how community structure influences ecosystem functions via functional traits. Formal causal analysis showed that biodiversity–ecosystem function correlations are noncausal associations. Using simulations, we showed how biodiversity correlations and species identity effects can arise from misspecification or incomplete mediation by functional trait composites. We also found that different types of model misspecification result in different patterns of residuals, which may be used to diagnose gaps in functional trait hypotheses. Treating the model misspecifications eliminated associations between species identity or biodiversity and ecosystem function. Finally, we provide an example of the analysis of field data to demonstrate how to use these insights to conduct a research program that has the goal of understanding the mechanistic trait relationships that link community structure to ecosystem function.Keywords
This publication has 43 references indexed in Scilit:
- Biodiversity loss and its impact on humanityNature, 2012
- Using Plant Functional Traits to Explain Diversity–Productivity RelationshipsPLOS ONE, 2012
- Substantial nutrient resorption from leaves, stems and roots in a subarctic flora: what is the link with other resource economics traits?New Phytologist, 2010
- Effect of thinning on LAI variance in heterogeneous forestsForest Ecology and Management, 2008
- The role of causal reasoning in understanding Simpson's paradox, Lord's paradox, and the suppression effect: covariate selection in the analysis of observational studiesEmerging Themes in Epidemiology, 2008
- Impacts of plant diversity on biomass production increase through time because of species complementarityProceedings of the National Academy of Sciences, 2007
- Representing general theoretical concepts in structural equation models: the role of composite variablesEnvironmental and Ecological Statistics, 2007
- EFFECTS OF BIODIVERSITY ON ECOSYSTEM FUNCTIONING: A CONSENSUS OF CURRENT KNOWLEDGEEcological Monographs, 2005
- Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversityOecologia, 1997
- The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations.Journal of Personality and Social Psychology, 1986