A comparative study of multiwavelength theoretical and observed light curves of Cepheid variables

Abstract

We analyse the theoretical light curves of Cepheid variables at optical (UBVRI) and near-infrared (JKL) wavelengths using the Fourier decomposition and principal component analysis methods. The Cepheid light curves are based on the full-amplitude, non-linear, convective hydrodynamical models for chemical compositions representative of Cepheids in the Galaxy (Y = 0.28, Z = 0.02), Large Magellanic Cloud (Y = 0.25, Z = 0.008) and Small Magellanic Cloud (Y = 0.25, Z = 0.004). We discuss the variation of light-curve parameters with different compositions and mass–luminosity levels as a function of period and wavelength, and compare our results with observations. For a fixed composition, the theoretical amplitude parameters decrease while the phase parameters increase with wavelength, similar to the observed Fourier parameters. The optical amplitude parameters obtained using canonical mass–luminosity Cepheid models exhibit a large offset with respect to the observations for periods between 7 and 11 d, when compared to the non-canonical mass–luminosity levels. The central minimum of the Hertzsprung progression for amplitude parameters shifts to the longer periods with decrease/increase in metallicity/wavelength for both theoretical and observed light curves. The principal components for Magellanic Cloud Cepheid models are consistent with observations at optical wavelengths. We also observe two distinct populations in the first principal component for optical and near-infrared wavelengths while the J band contributes to both populations. Finally, we take into account the variation in the convective efficiency by increasing the adopted mixing length parameter from the standard 1.5 to 1.8. This results in a zero-point offset in the bolometric mean magnitudes and in amplitude parameters (except close to 10 d), reducing the systematically large difference in theoretical amplitudes.