Quantitative analysis of carbon isotopic ratios in carbon stars: II. The effect of model atmosphere on the iso-intensity method

We discuss the analysis of ¹²C/¹³C ratios in cool carbon stars presented by de Laverny & Gustafsson (1998), who questioned the reliability of the iso-intensity method used by Ohnaka & Tsuji (1996). We show that the systematic discrepancy of ¹²C/¹³C ratios between Lambert et al. (1986) and Ohnaka & Tsuji (1996) cannot be attributed to the uncertainty of the iso-intensity method. The analysis of the iso-intensity method done by de Laverny & Gustafsson (1998) differs from that of Ohnaka & Tsuji (1996), defining the abscissa of curves of depth growth in a completely different manner. Namely, we derived the abscissa directly from model atmospheres, while they simply assumed a single excitation temperature whose value is never accurately derived. The high sensitivity of the iso-intensity method to model atmospheres, reported in their work, can be attributed to an incorrect definition of the abscissa of curves of depth growth. In fact, we show that the determination of ¹²C/¹³C ratios by the iso-intensity method is not so sensitive to model atmospheres (atmospheric structure itself and stellar parameters) as they claim, when the abscissa is properly calculated. In addition, we demonstrate that our model atmospheres can reproduce photometric and spectrophotometric observations fairly well. Therefore, their conclusion that the iso-intensity method is risky and unreliable for determining ¹²C/¹³C ratios in cool carbon stars cannot be justified.