Unveiling Galaxy Structures: Systematic Analysis of Bulge+Disk Decomposition Using Simulated JWST/NIRCam Observations

TL;DR

This study systematically assesses the accuracy of bulge+disk galaxy decomposition using simulated JWST/NIRCam images, providing insights into the reliability of structural measurements at high redshift.

Contribution

It introduces a comprehensive evaluation of single and double Sersic model fittings on simulated JWST data, establishing benchmarks for morphological analysis of distant galaxies.

Findings

Single Sersic fits recover total magnitudes within 0.5 mag for galaxies down to 27 mag.

Double Sersic models accurately recover bulge and disk parameters down to 26 mag.

Model parameter reliability improves with SNR exceeding 10.

Abstract

Characterizing and accurately decomposing galaxies into structural components, such as bulges and disks, is essential for understanding galaxy formation and evolution, particularly at high redshift, where galaxies are compact and faint. Leveraging the unparalleled resolution and sensitivity of JWST and imaging data from CEERs program, we simulate galaxies with bulge+disk components and assess the effectiveness of single and double S\'ersic model fittings, respectively. We first evaluate the performance of single S\'ersic fits, and find it can recover total magnitudes (i.e., within 0.5 mag), and size (i.e., within 0.2 dex), down to 27 mag. The features that emerged in the residual map can properly reflect the underlying two-component structures. We also show that S\'ersic indices can serve as proxies for the bulge-to-total flux ratio (B/T). For double S\'ersic models, we find comparable accuracy in recovering bulge and disk magnitudes (i.e., within 0.5 mag), and effective radius (i.e., within 0.2 dex), down to 26 mag. To quantitatively determine whether a double S\'ersic model better describes our two-component systems compared to a single S\'ersic profile, we evaluate the Bayesian Information Criterion for both model configurations. To extend the applicability of our results to other NIRCam programs, we evaluate the signal-to-noise ratio (SNR) of the simulated galaxies and find that model parameters are reliably reproduced when the SNR exceeds 10. Our work demonstrates the detailed morphological measurement uncertainties using single and double S\'ersic models, which provides an essential reference for future JWST/NIRCam-based morphological studies, especially for high-redshift galaxies.