Erratum: “Standing on the Shoulders of Dwarfs: The Kepler Asteroseismic LEGACY Sample. I. Oscillation Mode Parameters” (2017, ApJ, 835, 172)
Résumé
In this erratum, we provide corrected sets of r 01,10,02 difference ratio values and associated uncertainties, which were overestimated in the original paper (as noted by Roxburgh 2017) due to a missing trimming in the post-processing of the Markov chain Monte Carlo (MCMC) chains for these values. The typical reduction in the ratio uncertainties from performing the trimming is a factor of 10 (see Figure 3). Other parameters optimized in the peak-bagging (for instance, individual mode frequencies) are unaffected, as the trimming was performed for these in the original work (Lund et al. 2017). We also provide updated values for the n D 2 values of l=3 modes. We note that the values presented here, as with those presented in the original work, are obtained from a single peak-bagging procedure (see Lund et al. 2017 for details) and have yet to be verified by independent analyses using the same input power spectra. Examples of the updated tables from the original paper are given in Tables 1-3. We note that tables with individual mode parameters (Table 2) have been added for completeness, but the parameters in these tables are unchanged compared to the original paper. In addition to the corrected values mentioned above, we provide covariance matrices for the mode frequencies, frequency difference ratios (r 01,10,02), and second differences (n D 2) for the LEGACY sample (Lund et al. 2017), which were not published with the original work. The values provided by this erratum will be available in the online version of the paper. Figure 1. Comparison between ratio distribution of = r n 01, 25 (n » m 3090 Hz) for KIC 9139151 from the full (green) and properly thinned MCMC chains (black). The dashed red line (on top of the black curve) shows the distribution obtained by sampling from the reported frequency values and corresponding uncertainties (assuming that these are normally distributed and uncorrelated). The central peak is captured by both distributions, but the wide background signal representing the ratio prior has disappeared from the thinned chains. Dotted lines indicate the distribution medians; dashed lines bound the corresponding 68% highest probability density intervals.
Origine : Fichiers produits par l'(les) auteur(s)