Community Assembly Model Inference (CAMI)

Community Assembly Model Inference (CAMI) infers community assembly processes by simulating phenotypic similarity and repulsion to quantify environmental filtering and competitive exclusion in ecological communities.

Key Features:

• Modeling phenotypic similarity and repulsion: Incorporates models of phenotypic similarity and repulsion to simulate environmental filtering and competitive exclusion and to parameterize their strengths.
• Advanced statistical techniques: Employs random forests and approximate Bayesian computation (ABC) to differentiate between community assembly models and perform parameter inference.
• Uncertainty accounting: Accounts for uncertainty in model selection and parameter estimates.
• Parameter estimation: Estimates parameters that determine the strength of assembly processes.

Scientific Applications:

• Ecological mechanism inference: Applied to infer mechanisms driving community assembly from phenotypic and environmental data across ecosystems.
• Plant communities on lava flow islands: Demonstrated on plant communities on lava flow islands to investigate environmental filtering and species interactions.

Methodology:

Simulates community assembly by integrating phenotypic traits with environmental and competitive dynamics, uses random forests for model classification, and applies approximate Bayesian computation (ABC) for parameter inference while accounting for uncertainty.