The softness diagram for MaNGA star-forming regions: diffuse ionized gas contamination or local HOLMES predominance?
We explore the so-called softness diagram -- whose main function is to provide the hardness of the ionizing radiation in star-forming regions -- in order to check whether hot and old low-mass evolved stars (HOLMES) are significant contributors to the ionization within star-forming regions, as suggested by previous MaNGA data analyses. We used the code HCm-Teff to derive both the ionization parameter and the equivalent effective temperature (T*), adopting models of massive stars and planetary nebulae (PNe), and exploring different sets of emission lines in the softness diagram to figure out the main causes of the observed differences in the softness parameter in the MaNGA and CHAOS star-forming region samples. We find that the fraction of regions with a resulting T* > 60 kK, which are supposedly ionised by sources harder than massive stars, is considerably larger in the MaNGA (66%) than in the CHAOS (20%) sample when the [SII] $λλ$ 6716,6731 emission lines are used in the softness diagram. However, the respective fractions of regions in this regime for both samples are considerably reduced (20% in MaNGA and 10% in CHAOS) when the [NII] emission line at $λ$ 6584 is used instead. This may indicate that diffuse ionised gas (DIG) contamination in the lower resolution MaNGA data is responsible for artificially increasing the measured T* as opposed to there being a predominant role of HOLMES in the HII regions.