[Paper Review] Photometric entropy of stellar populations and related diagnostic tools

A. Buzzoni|arXiv (Cornell University)|Sep 20, 2005

Spectroscopy Techniques in Biomedical and Chemical Research3 citations

TL;DR

This paper introduces photometric entropy as a statistical diagnostic tool to quantify the effective number of luminous stellar contributors in unresolved stellar populations, such as globular clusters and distant galaxies. By linking photometric fluctuations to the discreteness and finiteness of stars, it enables robust estimation of population properties despite observational crowding and seeing limitations, with key results showing measurable color and spectral index scatter in high-S/N data that can constrain luminosity distance and age-metallicity degeneracy.

ABSTRACT

We discuss, from a statistical point of view, some leading issues that deal with the study of stellar populations in fully or partially unresolved aggregates, like globular clusters and distant galaxies. A confident assessment of the effective number and luminosity of stellar contributors can provide, in this regard, a very useful interpretative tool to properly assess the observational bias coming from crowding conditions or surface brightness fluctuations. These arguments have led us to introduce a new concept of "photometric entropy" of a stellar population, whose impact on different astrophysical aspects of cluster diagnostic has been reviewed here.

Motivation & Objective

• To address observational biases from crowding and surface brightness fluctuations in unresolved stellar systems such as globular clusters and distant galaxies.
• To develop a statistical framework that quantifies the effective number of luminous stellar contributors in stellar populations.
• To unify the treatment of photometric fluctuations arising from finite stellar discreteness across different observational conditions.
• To provide diagnostic tools for high-resolution photometry and spectroscopy that can break degeneracies in stellar population synthesis models.
• To enable distance measurements via statistical color scatter in high-S/N galaxy imaging.

Proposed method

• Defines photometric entropy as a statistical measure derived from the variance in integrated luminosity and color, reflecting the effective number of contributing stars.
• Uses Poisson statistics to model the distribution of stars across spatial or photometric cells, with expected fluctuations scaling as √N_eff.
• Applies the formula σ(B−V) ≤ √(σ²(B) + σ²(V)) to estimate statistical scatter in integrated color, dependent on the effective number of luminous contributors.
• Extends the approach to spectroscopy by modeling variance in narrow-band indices, such as Lick or UV indices, across isophotes.
• Uses population synthesis models (e.g., Buzzoni 1989, 1993) with updated UV-blue spectral libraries to compare age and HB morphology effects on N_eff and photometric entropy.
• Demonstrates that older, metal-poor populations with few bright HB stars exhibit higher photometric entropy and greater spectral variance than younger, main-sequence-dominated populations.

Experimental results

Research questions

• RQ1 How can photometric fluctuations in unresolved stellar systems be quantified as a function of the effective number of luminous contributors?
• RQ2 To what extent do surface brightness fluctuations and crowding effects arise from the same underlying physical cause—finite stellar discreteness?
• RQ3 Can statistical fluctuations in integrated color (e.g., B−V) be used to derive absolute luminosity and hence luminosity distance in high-S/N galaxy imaging?
• RQ4 How does photometric entropy vary between stellar populations of different ages and initial mass functions, and what does this imply for population synthesis modeling?
• RQ5 Can spectral variance in narrow-band indices serve as a diagnostic tool to break the age-metallicity degeneracy in stellar populations?

Key findings

• Photometric entropy provides a robust statistical measure of the effective number of luminous contributors in unresolved stellar populations, with higher entropy indicating greater statistical scatter due to fewer, brighter stars.
• For a typical 15 Gyr stellar population with a blue horizontal branch, the effective number of contributors (N_eff) is significantly lower than for a 2 Gyr population, leading to higher photometric entropy.
• In high-S/N galaxy imaging (S/N ≫ 100), a measurable color scatter of ~0.01 mag in B−V is expected for a 10¹¹ L⊙ elliptical galaxy at 15 Mpc, with a pixel sampling of ~10⁵ L⊙/px.
• Theoretical models show that despite nearly identical spectral energy distributions, a 15 Gyr SSP with blue HB stars exhibits much larger variance in spectral features than a 2 Gyr SSP with red HB, due to fewer dominant contributors.
• Spectral variance in narrow-band indices (e.g., Lick or UV indices) is expected to be about twice as large in older populations, providing a diagnostic tool to distinguish age and metallicity effects.
• The statistical scatter in integrated color is bounded by |σ(B)−σ(V)| ≤ σ(B−V) ≤ √(σ²(B) + σ²(V)), with the upper limit achievable only under perfect positive correlation between B and V luminosity contributors.