Erratum

Abstract

In the originally published version of this manuscript, an indexing issue in our analysis code caused certain properties to be misassigned to incorrect objects in our sample. This occurred during three separate parts in our analysis: calculating the rest-frame wavelengths in which each object was observed λrest, calculating the individual and ensemble structure functions (SFs), and calculating the intrinsic variability of individual objects σ0. This error caused the incorrect redshift z to be used for certain objects when computing λrest and the incorrect light curves to be used when calculating SFs and σ0. All other analyses are consistent with the correctly assigned redshifts and light curves, including the power spectrum density (PSD) analysis. It should be noted that the reported redshifts and light curves for the sample are correct, but were used incorrectly due to this bug in code. This error affected certain columns in the output data products, located at https://zenodo.org/record/7624056. In particular, the SIG0 x, SIG0 x ERR, LAMBDA REST x, DT REST x, SF x, SF x ERR L, and SF x ERR U columns were incorrect in the data table described in Table 1 in the original manuscript, where ‘x’ signifies the band, and can be either g, r, or i. Similarly, the columns DT REST x, SF x, and SF x ERR for the different ensembles in the data table described in Table 2 in the original manuscript were also incorrect. These values have been corrected using the correct analysis and reuploaded to the same location. This error affected a number of figures in the original manuscript, namely Figs 2, 8, 9, and 10. All updated figures are shown here, with the following changes to our results: (i) The values for σ0 are now accurate, and follow all prior relations discussed in the original manuscript. (ii) The dependence of τDRW, rest on λrest is weaker, which better matches the results of MacLeod et al. (2010), with an updated relation of τDRW, rest ∝ λ0.35 ± 0.14 for the subsample of 27 objects with τDRW rest shorter than 20 per cent of the baseline. (iii) The subsample still displays a weak dependence between SF∞ and λrest, though deviating further from the results in MacLeod et al. (2010). (iv) The total sample produces a stronger dependence than in the original manuscript, with τDRW, rest ∝ λ0.43 ± 0.10. The strengthened dependence of τDRW, rest on λrest for the entire sample is still consistent with the results found in MacLeod et al. (2010). (v) The total sample produces a weaker dependence with respect to SF∞, with SF∞ ∝ λ0.00 ± 0.07 . (vi) After reconstructing the SFs for individual objects, the correct ensemble SFs are displayed in Figs 9 and 10. In Fig. 9, the subsample SFs more closely match the total ensemble SF. Fig. 10 shows similar results to the original figure, with slightly differing values for each of the subsamples shown. In summary, this erratum fixed an indexing error that affected a few figures and tabulated data in Table 1. However, none of the main conclusions are changed. Show more