Erratum: Nucleon form factors and root-mean-square radii on a (10.8 fm)4 lattice at the physical point (Physical Review D (2019) 99 (014510) DOI: 10.1103/PhysRevD.99.014510)

In our analysis we employed different normalization factors from Eqs. (20)–(23) in the paper to evaluate the electric, magnetic, axial-vector, induced pseudoscalar, and pseudoscalar form factors in the nonzero q2. In some parts of the normalization factors, MN was used instead of EN. We corrected the normalization factors in our analysis code, and then recalculated the form factors with the correct normalization factors as in the equations. The effect of this correction increases as q2 increases. As a result, GE(q2) and FA(q2) are overestimated from the experiments in the large q2 region as presented in Figs. 3, 4, and 14. The root-mean-square (RMS) charge radius and the axial-vector RMS radius become smaller than the ones quoted in the paper, and are underestimated from the experiments as shown in Figs. 5 and 15, respectively. In contrast to these RMS radii, other results obtained from the magnetic, induced pseudoscalar, and pseudoscalar form factors are reasonably consistent with the ones that appeared in the paper. Figure 23 presents that the RMS charge radius becomes reasonably consistent with the recent lattice results [6–9]. Thus, the finite volume effect could be small in this radius, although we discussed the possibility of its existence in the paper. For the charge radius, it is important to understand the reason of the discrepancy between the lattice results and the experiments for solving the proton size puzzle. Towards understanding of the discrepancy, we plan to measure the RMS charge radius with a finer lattice spacing in the next calculation to estimate discretization error, which was not estimated in the paper. Our best estimated values for the RMS radii and magnetic moments in Eqs. (29)–(32) of the paper are replaced by the following values, (Formula Presented) (Figure Presented).