Predictions of the community assemblage in a temperate forest through indicators that evaluate the anthropogenic disturbance effect on natural regeneration

Abstract The causal pathways of the anthropogenic disturbance effect on the above-ground vegetation, seed bank, and seed rain can provide information to understand the dynamic of natural regeneration in order to predict changes in the community assemblage. However, due to the spatial and temporal complexity of the natural regeneration process, these pathways have rarely been tested empirically. Therefore, to understand these pathways on natural regeneration, we propose the integration of composition, richness and functional diversity as indicators of assemblage vulnerability, and apply them to the above-ground vegetation, seed bank, and seed rain. In an Abies religiosa forest, in Mexico City, we determined anthropogenic disturbance indicators, environmental indicators and indicators of assemblage vulnerability. In order to assess if anthropogenic disturbances increase the abundance of weeds and modify functional diversity, we built linear mixed models (LMMs), generalized linear mixed models (GLMMs) and structural equation models (SEMs) and selected the best-fitting models according to Akaike's information criterion. We carried out RLQ/Fourth-corner tests in order to determine those regeneration functional traits of weeds that favor their tolerance to anthropogenic disturbances. We found that the anthropogenic disturbances could drive the community into an alternative stable state, in which the presence of weeds could homogenize its functional diversity. Our results showed that the applied indicators reduce the complexity of the study of the natural regeneration process and determine the causal pathways of the anthropogenic disturbance, in order to predict the trends of the community assemblage.