Erratum for: A&A, 650, A103 (2021), https://doi.org/10.1051/0004-6361/202039402

Table 3 in the original paper reports inaccurate values of $Q_{\mathrm{e}\mathrm{x}\mathrm{p},\mathrm{Y}\mathrm{S}\mathrm{C}}^0$ and $Q_{\mathrm{o}\mathrm{b}\mathrm{s}}^0$ (and hence of the related quantities Q_exp, tot, Q_exptot/Q_obs, and f_esc). The error in $Q_{\mathrm{e}\mathrm{x}\mathrm{p},\mathrm{Y}\mathrm{S}\mathrm{C}}^0$ results from amistake in the sampling of the cluster probability distribution functions (PDFs), whereas $Q_{\mathrm{o}\mathrm{b}\mathrm{s}}^0$ was mistakenly computed from non-dereddened values of L(Hα). In Table 1 we provide a revised version of the original Table 3, and in Figs. 1 and 2 we show revised versions of the original Figs. 10 and 11.

Fig. 1. Revised version of Fig. 10 in the original paper.

Fig. 2. Revised version of Fig. 11 in the original paper.

The revised values do not change any of the conclusions stated in the original paper. In the following we provide revised text for the affected sections.

7. Ionisation budget

7.4. Resulting budget

We observe that in regions 2, 3, 6, and 7, the value of f_esc (with the related uncertainty) reaches unphysical values f_esc <  0, indicating that either the models are underestimating the photon flux, or the observed luminosity is being overestimated. This is possibly due to the reddening correction or to the exact location of the region boundaries.

Overall, we find $f_{\mathrm{esc},\mathrm{H}\mathrm{I}\mathrm{I}}={0.43}_{-0.23}^{+0.13}$ for the entire population of H II regions. We observe that the stellar population in the diffuse ionised gas (DIG) produces a more than sufficient amount of ionising photons (Q(H⁰)_exp >  Q(H⁰)_obs), and that the DIG is therefore consistent with being self-ionised, with f_esc, DIG = 0.85${}_{-0.07}^{+0.04}$. This holds also if considering a maximum mass of 30 M_⊙ for field O stars, in which case we find f_esc, DIG = 0.50${}_{-0.20}^{+0.12}$. Our conclusion therefore remains unchanged: in our field of view (FoV), we observe that the sources of ionising photons produce a photon flux that is more than sufficient to explain the emission of the ionised interstellar medium (ISM), both within and outside the H II regions.

7.5. Escape fraction from individual H II regions

Figures 1 and 2 are revised versions of Figs. 10 and 11 in the original paper. Our finding still stands: we do not observe a trend between f_esc and the region morphology nor between f_esc and the age of the stellar population in the region.

8. Discussion

Our revised estimate of the fraction of diffuse gas expected is

$$f_{\mathrm{DIG},\exp }=\frac{Q_{\exp ,\mathrm{DIG}}^0}{Q_{\exp ,\mathrm{DIG}}^0+Q_{\exp ,\mathrm{H}\mathrm{II}}^0}\gtrsim 0.15.$$

This is in even closer agreement with Paper I.

The revised Q(H⁰) values still point to a DIG consistent with being self-ionised by field stars and clusters, with an overabundance of ionising photons f_esc∼ 0.85${}_{-0.07}^{+0.04}$; we confirm this trend also when considering our lower-limit estimate. Overall the H II regions are leaking ionising photons at a rate f_esc∼ 0.43${}_{-0.23}^{+0.13}$, with five out of eight having an f_esc ≳ 0.3. Even when considering the uncertainties, we find f_vesc >  0 for the DIG overall and for four out of the eight H II regionsinspected.

9. Conclusions

Overall, we find an escape fraction f_esc = 0.43${}_{-0.23}^{+0.13}$ for the population of H II regions, and that the DIG in our FoV is more than consistent with being self-ionised, with an f_esc = 0.85${}_{-0.07}^{+0.04}$. This holds even when considering a lower-limit estimate for the DIG flux, which was derived by assuming a maximum mass of 30 M_⊙ for the field O stars. We furthermore find that the f_DIG, exp ≳ 0.15 obtained by modelling the DIG stellar population is in good agreement with the DIG fraction derived from the observed Hα luminosity in Paper I, f_DIG, obs = 0.15. Finally, we observe an f_esc ≳ 0.3 in five out of the eight studied regions.

© Della Bruna et al. 2022

All Tables

All Figures

	Fig. 1. Revised version of Fig. 10 in the original paper.
In the text

	Fig. 2. Revised version of Fig. 11 in the original paper.
In the text