[Paper Review] A uniform set of optical/NIR photometric zero points to be used with CHORIZOS

J. Maíz Apellániz|arXiv (Cornell University)|Sep 15, 2006

Satellite Image Processing and Photogrammetry6 citations

TL;DR

This paper presents a uniform, high-accuracy set of photometric zero points for six optical and near-infrared photometric systems—Johnson UBV, Cousins RI, Strömgren uvby, Tycho-2 BT VT, 2MASS JHKs, and SDSS ugriz—calibrated against a new, improved Vega spectrum. The zero points, derived from HST/STIS spectrophotometry and literature data, enable precise synthetic photometry in the Bayesian code CHORIZOS, reducing systematic errors to ≤1% and improving the accuracy of stellar parameter estimation from multi-filter photometry.

ABSTRACT

I have recently combined HST/STIS spectrophotometry with existing photometric data to analyze the calibration of three standard optical photometry systems: Tycho-2 B_TV_T, Stromgren uvby, and Johnson UBV. In this contribution I summarize those results, present new ones for 2MASS JHK_s, and combine them with recent literature results to generate a uniform set of zero points for six photometric systems, the above mentioned plus Cousins RI and SDSS ugriz. With the exception of the latter system, the zero points use the new Vega spectrum presented at this meeting by Ralph Bohlin. I also discuss the implementation of these results in CHORIZOS, a Bayesian photometric code that compares multi-filter observational data with spectral energy distributions to solve the inverse problem of finding the models which are compatible with the observations.

Motivation & Objective

• To resolve systematic errors in photometric zero points that degrade the accuracy of inverse photometric solutions in multi-filter stellar population modeling.
• To unify disparate photometric systems (e.g., Johnson, Cousins, SDSS, 2MASS) under a common, consistent zero-point framework using a modern Vega spectrum.
• To improve the accuracy of CHORIZOS, a Bayesian photometric code, by replacing outdated zero points with new, systemically calibrated values.
• To enable high-precision comparison between observed photometry and synthetic spectral energy distributions (SEDs) by minimizing systematic offsets in magnitude systems.
• To support large-scale surveys by providing stable, transferable zero points that remain valid even when reference SEDs are updated.

Proposed method

• The zero points are computed using Eq. (1), which relates observed magnitudes to synthetic magnitudes via filter passbands and SEDs, with the reference SED being the new Vega spectrum from Bohlin.
• The method combines HST/STIS spectrophotometry of stars with existing photometric data to derive system-specific zero points (ZP r,p) for each filter, ensuring consistency across systems.
• For systems not based on AB or ST standards (e.g., Johnson+Vega), non-zero zero points are calculated to align synthetic magnitudes with observed values, preserving published data.
• The approach avoids recalibrating observed magnitudes by instead adjusting synthetic magnitudes through non-zero ZP values, maintaining data integrity.
• The zero points are validated by comparing synthetic colors from CHORIZOS with observed data on low-extinction OB stars, showing agreement within 1%.
• The results are integrated into CHORIZOS version 2.1.3, enabling improved multi-filter SED fitting with full Bayesian inference and future support for spectrophotometry and detection limits.

Experimental results

Research questions

• RQ1 What is the systematic offset between observed and synthetic magnitudes when using outdated or inconsistent photometric zero points in multi-filter photometry?
• RQ2 How can a uniform, high-accuracy zero-point system be established across multiple optical and near-infrared photometric systems?
• RQ3 To what extent do updated zero points based on HST/STIS spectrophotometry reduce systematic errors in synthetic photometry?
• RQ4 Can the inclusion of new, precise zero points in CHORIZOS significantly improve the accuracy of stellar parameter estimation from photometric data?
• RQ5 How does the use of a modern, high-fidelity Vega spectrum impact the consistency and accuracy of photometric zero points across different systems?

Key findings

• The new zero points for Johnson UBV, Cousins RI, Strömgren uvby, Tycho-2 BT VT, 2MASS JHKs, and SDSS ugriz are derived with an accuracy of ≤1%, significantly improving consistency across systems.
• The zero points are based on a new, high-resolution Vega spectrum from Ralph Bohlin, reducing uncertainties in the reference SED and minimizing systematic offsets.
• The implementation of these zero points in CHORIZOS version 2.1.3 reduced systematic errors in synthetic photometry to ≤1%, as confirmed by testing on low-extinction OB stars.
• The method preserves published observed magnitudes while correcting synthetic magnitudes via non-zero ZP values, enabling backward compatibility with historical data.
• Preliminary validation shows that synthetic magnitudes computed with the new zero points match observed data within 0.01 mag, confirming the 1% accuracy target.
• Future versions of CHORIZOS will support spectrophotometry, multicolor indices, detection limits, and full Bayesian priors, with potential minor zero-point updates based on new data.