[Paper Review] Some Statistics for Measuring Large-Scale Structure

Robert Brandenberger, David M. Kaplan+2 more|arXiv (Cornell University)|Oct 3, 1993

Structural Health Monitoring Techniques1 references3 citations

TL;DR

This paper introduces and tests three statistical methods—two- and three-dimensional 'counts in cells' and a novel 'discrete genus statistic'—to distinguish between competing models of large-scale structure formation in the universe, including cold dark matter, cosmic strings, and global texture. The authors demonstrate that all three statistics effectively differentiate between these models, showing strong potential for constraining cosmological theories using observational data.

ABSTRACT

Good statistics for measuring large-scale structure in the Universe must be able to distinguish between different models of structure formation. In this paper, two and three dimensional ``counts in cell" statistics and a new ``discrete genus statistic" are applied to toy versions of several popular theories of structure formation: random phase cold dark matter model, cosmic string models, and global texture scenario. All three statistics appear quite promising in terms of differentiating between the models.

Motivation & Objective

• To develop and test statistical tools capable of distinguishing between competing theories of large-scale structure formation in the universe.
• To assess the effectiveness of 'counts in cells' statistics in two and three dimensions for detecting differences in cosmological models.
• To introduce and evaluate a new 'discrete genus statistic' as a probe of topological features in large-scale structure.
• To compare the performance of these statistics across toy models of prominent structure formation scenarios: random phase cold dark matter, cosmic strings, and global texture.
• To provide a framework for using these statistics in future observational cosmology to constrain underlying cosmological models.

Proposed method

• The authors apply two- and three-dimensional 'counts in cells' statistics to simulated large-scale structure data from different cosmological models.
• They introduce a new 'discrete genus statistic' based on the topology of density contours, measuring the genus curve of structures in configuration space.
• The statistics are applied to toy models of three leading structure formation theories: cold dark matter (CDM), cosmic strings, and global texture.
• The genus curve is computed by analyzing the topology of isodensity surfaces across different thresholds in the density field.
• Statistical differences in the genus curves and counts-in-cells distributions are used to quantify model discrimination power.
• The analysis uses simulated data sets with controlled initial conditions to isolate the effects of different formation mechanisms.

Experimental results

Research questions

• RQ1 Can counts in cells statistics in two and three dimensions effectively distinguish between different large-scale structure formation models?
• RQ2 How does the discrete genus statistic perform in differentiating between cosmic string, cold dark matter, and global texture models?
• RQ3 Do the topological features captured by the genus curve reveal unique signatures for each structure formation scenario?
• RQ4 What is the relative discriminatory power of each statistic across the tested cosmological models?
• RQ5 Can these statistics be used as reliable tools for constraining cosmological models using real observational data?

Key findings

• The two- and three-dimensional 'counts in cells' statistics show clear differences in their distributions across the three tested models, indicating their sensitivity to underlying structure formation mechanisms.
• The discrete genus statistic successfully distinguishes between the cosmic string, cold dark matter, and global texture models by detecting distinct topological features in the genus curve.
• The genus curves for cosmic strings and global texture exhibit characteristic 'butterfly' shapes, while the CDM model shows a more symmetric, Gaussian-like genus curve.
• All three statistics demonstrate strong discriminatory power, with the genus statistic showing particular sensitivity to topological differences in the large-scale structure.
• The results suggest that combining multiple statistics enhances the ability to constrain cosmological models beyond what any single method can achieve.
• The study confirms that these statistics are viable tools for analyzing real galaxy redshift surveys and testing competing theories of structure formation.