Detecting non-neutral modules in species co-occurrence data: principles and application to plant communities

Area of Science:

• Community Ecology
• Theoretical Ecology
• Biodiversity Research

Background:

• Inferring ecological assembly processes from species co-occurrence data is challenging.
• Existing methods analyzing spatial covariance are limited by non-random patterns arising from various factors.
• Process-based null hypotheses are crucial for accurate inference.

Purpose of the Study:

• To develop a novel framework for inferring community assembly processes using neutral theory.
• To utilize the concept of 'rank consistency' in species co-occurrences as a key indicator.
• To identify species groups exhibiting non-neutral co-occurrence patterns.

Main Methods:

• Developed pairwise tests for species rank consistency along ecological gradients.
• Constructed a species network connecting rank-consistent species.
• Identified network modules representing groups with non-neutral co-occurrences.
• Interpreted modules by relating species composition to functional traits.

Main Results:

• Successfully retrieved pre-defined functional groups from virtual data without false positives.
• Identified ecologically meaningful modules in tropical tree and Mediterranean herbaceous communities.
• Demonstrated the framework's potential for understanding assembly processes.

Conclusions:

• Species rank consistency provides a robust approach to inferring community assembly.
• Network modules derived from rank consistency highlight deviations from neutral theory.
• This framework offers a significant step towards understanding ecological assembly processes.