Estimating dispersal from short distance spatial autocorrelation.

Area of Science:

• Population genetics
• Spatial statistics
• Ecological modeling

Background:

• Theoretical studies link population spatial statistics to dispersal measures.
• Dispersal can be indirectly estimated from genetic variation patterns in populations.
• Short-distance spatial correlations are robust and statistically powerful for dispersal estimation.

Purpose of the Study:

• To integrate theoretical results for robust and flexible estimation of dispersal.
• To provide practical, broad empirical guidelines for estimating dispersal.
• To focus on using short-distance autocorrelation for estimating neighborhood size and dispersal variance.

Main Methods:

• Review and synthesis of theoretical studies on spatial statistics and dispersal.
• Application of Moran's I-statistic for diploid genotypes converted to allele frequencies.
• Comparison and extension to other statistical approaches for dispersal estimation.

Main Results:

• Developed a method to indirectly estimate dispersal from spatial genetic variation.
• Short-distance autocorrelation is highlighted as a robust and powerful estimator.
• Provided empirical guidelines for practical application in plant and animal populations.

Conclusions:

• Spatial genetic patterns provide a viable method for estimating dispersal.
• Short-distance autocorrelation offers a flexible and robust approach.
• The developed guidelines enhance the practical application of these methods in ecological research.