Competitive exclusion, or species aggregation? : An aid in deciding

Area of Science:

• Ecology
• Community Ecology
• Biogeography

Background:

• A long-standing debate exists on whether species co-occurrence data, particularly from archipelagos, can reveal the ecological forces structuring biological communities.
• Analyzing species distributions is crucial for understanding ecological interactions and community assembly processes.

Purpose of the Study:

• To develop and apply novel quantitative scores (S, C, T) to analyze species co-occurrence patterns and test hypotheses about community structure.
• To investigate the role of interspecific interactions, such as competition, in shaping species distributions on islands.

Main Methods:

• Introduced three quantitative scores: S (shared islands), C (checkerboard score for mutual exclusion), and T (togetherness score for aggregation).
• Utilized null hypothesis testing by generating constrained simulated incidence patterns to compare with empirical data.
• Focused analysis on confamilial species subsets within the Vanuatu archipelago to identify potential competition dynamics.

Main Results:

• Demonstrated that high checkerboard scores (C) can arise from aggregation forces, not just avoidance, due to inherent constraints in presence/absence matrices.
• Empirical data from Vanuatu showed confamilial species share more islands and exhibit less checkerboardedness (more aggregation) than simulated patterns.
• Observed species distributions significantly deviate from random colonization, indicating non-random assembly processes.

Conclusions:

• The findings suggest that while competition may exist, its effects on community structure are potentially overcome by other ecological factors in the studied system.
• The developed scores and simulation methods provide a robust framework for analyzing community structure and inferring ecological interactions.
• The approach is broadly applicable to diverse ecological problems investigating the forces driving community assembly.