Single Stellar Populations in the Near-Infrared - I. Preparation of the IRTF spectral stellar library

TL;DR

This paper thoroughly characterizes the IRTF stellar library, assessing its spectral quality, calibration, and stellar parameters to establish it as a reliable resource for near-infrared stellar population modeling.

Contribution

It provides a detailed analysis of the IRTF spectral library's stellar parameters, flux calibration, and resolution, confirming its suitability for stellar population models.

Findings

Good broad-band flux calibration confirmed

Resolution increases from 6 to 10 angstroms across the NIR range

Library is suitable for use in stellar population modeling

Abstract

We present a detailed study of the stars of the IRTF spectral library to understand its full extent and reliability for use with Stellar Population (SP) modeling. The library consist of 210 stars, with a total of 292 spectra, covering the wavelength range of 0.94 to 2.41 micron at a resolution R = 2000. For every star we infer the effective temperature (Teff), gravity (logg) and metallicity ([Z/Zsun]) using a full-spectrum fitting approach in a section of the K band (2.19 to 2.34 micron) and temperature-NIR colour relations. We test the flux calibration of these stars by calculating their integrated colours and comparing them with the Pickles library colour-temperature relations. We also investigate the NIR colours as a function of the calculated effective temperature and compared them in colour-colour diagrams with the Pickles library. This latter test shows a good broad-band flux calibration, important for the SP models. Finally, we measure the resolution R as a function of wavelength. We find that the resolution increases as a function of lambda from about 6 angstrom in J to 10 angstrom in the red part of the K-band. With these tests we establish that the IRTF library, the largest currently available general library of stars at intermediate resolution in the NIR, is an excellent candidate to be used in stellar population models. We present these models in the next paper of this series.