As we know, Type Ia supernovae are used as standard candles as their absolute luminosities are expected to be roughly constant. The consensus model, as seen from the Wikipedia page, suggests this is because Type Ia SNe come from a binary star system with a white dwarf and a red giant, where the mass of the red giant slowly accretes onto the white dwarf, until the Chandrasekhar limit where runaway thermonuclear fusion begins, causing the supernova.

However, a paper linked in the Wikipedia page suggests that the above "single degenerate model" only accounts for up to 20% of Type Ia supernova. As far as I understand, the main alternative to the single degenerate scenario is a double degenerate scenario, i.e. the merger of two white dwarves. Of course, the double degenerate scenario is not constrained by the Chandrasekhar limit, and consequently do not have a fixed luminosity.

Question. If only <20% of Type Ia SNe are single-degenerate and have a fixed luminosity, then how can we use them as standard candles at all?

I get the impression from the above paper that it is in general very difficult to understand whether a particular Type Ia supernova fits into the single or double degenerate scenario. In this case, how can we have any confidence in distance measurements with Type Ia SNe? I would appreciate if an expert can enlighten me in this regard. Thanks in advance!

As we know, Type Ia supernovae are used as standard candles since their absolute luminosities are expected to be roughly constant. The consensus model, as seen from the Wikipedia page, suggests this is because Type Ia SNe come from a binary star system with a white dwarf and a red giant, where the mass of the red giant slowly accretes onto the white dwarf, until it reaches the Chandrasekhar limit, where runaway thermonuclear fusion begins, causing the supernova.

However, a paper linked in the Wikipedia page suggests that the above "single degenerate model" only accounts for $\leq 20$% of Type Ia supernova. As far as I understand, the main alternative to the single degenerate scenario is a double degenerate scenario, i.e. the merger of two white dwarfs. Of course, the double degenerate scenario is not constrained by the Chandrasekhar limit, and consequently does not have a fixed luminosity.

Question. If only <20% of Type Ia SNe are single-degenerate and have a fixed luminosity, then how can we use them as standard candles at all?

I get the impression from the above paper that it is in general very difficult to understand whether a particular Type Ia supernova fits into the single or double degenerate scenario. In this case, how can we have any confidence in distance measurements with Type Ia SNe?