SIMBAD references

We present a theoretical investigation of the effect of multiple ionization sources in Hii regions on the total elemental abundances derived from the analysis of collisionally excited emission lines. We focus on empirical methods based on direct temperature measurements that are commonly employed in cases when the temperature of the nebular gas can be determined from the ratio of nebular to auroral lines of (e.g.) doubly ionized oxygen. We find that direct temperature methods that employ a two-temperature zone approach (DT2T methods) are very robust against the spatial distribution of sources. Errors smaller than 0.15 dex are estimated for regions where the metallicity is twice solar and errors below 0.05 dex for solar metallicities and below. The biases introduced by the spatial distribution of the ionization sources are thus much smaller for DT2T methods than for strong line methods, previously investigated by Ercolano, Bastian & Stasi{nacute}ska. Our findings are in agreement with the recent study of Hii regions in NGC 300 by Bresolin et al.