A neutral sampling formula for multiple samples and an 'exact' test of neutrality

Area of Science:

• Ecology
• Biodiversity Science
• Theoretical Ecology

Background:

• Neutral theory is a key framework for understanding biodiversity.
• Existing sampling formulas are limited to single local samples, hindering analysis of spatially separated data.
• There's a need for advanced tools to assess neutral processes like dispersal limitation and stochasticity.

Purpose of the Study:

• To develop a generalized sampling formula for multiple, spatially separated samples from a single metacommunity.
• To introduce a novel, model-free test for ecological neutrality.
• To evaluate the applicability of neutral theory to neotropical forest tree communities.

Main Methods:

• Developed a generalized sampling formula for multiple spatial samples.
• Created an algorithm for generating neutral model data sets.
• Implemented a neutrality test comparing observed data probability with model-generated data probability.
• Applied the test to tree abundance data from three Panamanian forest plots.

Main Results:

• The new sampling formula accommodates multiple, spatially separated samples.
• When using plot-specific parameters, the neutral model could not be rejected for the Panamanian tree data.
• When using previously reported parameters (BCI), the neutral model was strongly rejected.
• The neutral model failed to simultaneously explain community structure at local and regional scales.

Conclusions:

• The developed tools offer improved analysis of biodiversity data within a neutral theory framework.
• Neutral theory may not fully capture the complex dynamics of the studied Panamanian tree communities across scales.
• Spatially implicit assumptions of the neutral model might be a limitation.