Uncertainties in the calibrations of star formation rate

Abstract

The calibrations of star formation rate (SFR) are prone to be affected by many factors, such as metallicity, initial mass function (IMF), evolutionary population synthesis (EPS) models and so on. In this paper we will discuss the effects of binary interactions, metallicity, EPS models and IMF on several widely used SFR calibrations based on the EPS models of Yunnan with and without binary interactions, BC03, SB99, PEGASE and POPSTAR. The inclusion of binary interactions makes these SFR conversion coefficients smaller (less than 0.2dex), and these differences increase with metallicity. The differences in the calibration coefficient between SFR and the luminosity of $\rm H\alpha$ recombination line (C$_{\rm H\alpha}$) and that between SFR and the ultraviolet (UV) fluxes at 1500 and 2800\,$\rm \AA$ (C$_{i, {\rm UV}}$), caused by IMF, are independent of metallicity (0.03-0.33\,dex) except $\Delta$C$_{\rm H\alpha, IMF}$ when using the POPSTAR and $\Delta$C$_{i, {\rm UV, IMF}}$ when using the PEGASE models. Moreover, we find that $L_{\rm 2800}$ is not suitable to the linear calibration of SFR at low metallicities. At last, we compare the effects of these several factors on the SFR calibrations considered in this paper. The effects of metallicity/IMF and EPS models on the C$_{\rm H\alpha}$ and C$_{\rm FIR}$ (the conversion coefficient between SFR and the far-infrared flux) are the largest among these factors, respectively. For the calibration between SFR and C$_{i, {\rm UV}}$, the effects of these several factors are comparable.