A new probabilistic method for quantifying n-dimensional ecological niches and niche overlap

Area of Science:

• Ecology
• Quantitative Biology
• Statistical Modeling

Background:

• Stable isotope analysis is crucial for understanding trophic relationships.
• Existing statistical tools for niche quantification have limitations in dimensionality and overlap estimation.

Purpose of the Study:

• To develop a novel probabilistic method for defining niche regions and estimating pairwise niche overlap.
• To extend niche analysis beyond two dimensions and provide directional overlap estimates.

Main Methods:

• Proposed a probabilistic method defining niche region (NR) as a probability region in multivariate space.
• Calculated overlap as the probability of an individual from one species within another's NR.
• Incorporated uncertainty using a Bayesian framework, independent of sample size.

Main Results:

• The method provides consistent and unique bivariate projections of multivariate niche data.
• Demonstrated application with three-dimensional stable isotope data.
• The approach is adaptable to any continuous ecological niche indicator in multiple dimensions.

Conclusions:

• This probabilistic method advances the quantification and comparison of ecological niches.
• The approach aligns better with Hutchinson's concept of the n-dimensional hypervolume.
• Offers a more robust framework for ecological niche analysis.