Joint species distribution modelling with the r-package Hmsc

Area of Science:

• Community Ecology
• Statistical Ecology
• Computational Ecology

Background:

• Joint Species Distribution Modelling (JSDM) is a powerful statistical approach for analyzing ecological community data.
• The Hierarchical Modelling of Species Communities (HMSC) framework offers a flexible method for fitting JSDMs, integrating environmental, trait, phylogenetic, and spatio-temporal data.
• Previous HMSC implementations were limited to MATLAB users, restricting broader accessibility.

Purpose of the Study:

• To introduce Hmsc 3.0, a user-friendly R implementation of the HMSC framework.
• To make advanced JSDM analyses accessible to a wider ecological research community.
• To provide a versatile tool for exploring community assembly processes using observational data.

Main Methods:

• Development and release of Hmsc 3.0, an R package for fitting JSDMs.
• Application of Hmsc 3.0 to diverse case studies using real and simulated ecological data (e.g., bird counts, environmental covariates, species traits, phylogenetic relationships).
• Demonstration of model fitting, parameter estimation, convergence diagnostics, model evaluation, and prediction capabilities for various data types (normal, count, presence-absence).

Main Results:

• Hmsc 3.0 successfully implements the HMSC framework in R, enabling JSDM fitting and post-processing.
• The package facilitates the estimation of species responses to environmental factors, considering species traits and residual associations.
• Demonstrated flexibility in handling different data distributions and incorporating various random effects structures.

Conclusions:

• Hmsc 3.0 significantly enhances the accessibility of advanced JSDM techniques for ecologists.
• The package provides a comprehensive and user-friendly platform for ecological community analysis.
• Facilitates deeper insights into community assembly and species interactions from observational data.