PS 56-105
Does species richness drive community production or vice versa? A causal analysis

Thursday, August 14, 2014
Exhibit Hall, Sacramento Convention Center
Donald R. Schoolmaster Jr., Wetland and Aquatic Research Center, U.S. Geological Survey, Lafayette, LA
James B. Grace, U.S. Geological Survey Wetland and Aquatic Research Center, Lafayette, LA
Background/Question/Methods

The causal relationships between biomass production and species richness have been an issue at the center of community ecology for over twenty years. Initially, models focused on proposing mechanisms whereby high levels of production could lead to limits on richness. Beginning in the 1990s, models began to be developed that proposed various mechanisms whereby richness might affect production. More recently, attempts have been made to derive theoretical expectations for the interconnections between production and richness using resource competition models within a metacommunity framework.  While all these theoretical efforts focus on the question of how productivity and richness may influence each other, a formal causal analysis of the problem has not been conducted. Our causal analyses focuses on 1) how dynamic models can be translated to causal diagrams to allow for formal causal analysis and (2) an evaluation of the causal claims based on the published literature.

Results/Conclusions

Our analyses show that causal statements of the form “X causes Y” can often be evaluated directly from the causal diagram derived from the dynamic model. For resource competition models, we evaluated the claims that (1) resource supply controls species richness, (2) community biomass does not exert a causal effect on species richness, and (3) species richness affects productivity. Our causal analyses show that within these models, and in opposition to previous claims, both resource supply and community biomass can exert causal influences on species richness, and that species richness has a causal effect on productivity. Simulations of the dynamic system confirm the conclusions of the causal analysis.