# A Hubble Space Telescope survey for novae in M87. III. Are novae good standard candles 15 days after maximum brightness?

## TL;DR

This study analyzes Hubble Space Telescope observations of M87 novae, demonstrating that novae with decline times over 10 days can serve as reliable standard candles approximately two weeks after maximum brightness, despite large scatter in earlier relations.

## Contribution

It provides evidence supporting the modified Buscombe - de Vaucouleurs hypothesis that novae with t2 > 10 days have nearly uniform brightness 15 days after maximum, improving distance measurement methods.

## Key findings

- Novas with t2 > 10 days have consistent magnitudes around 15 days post-maximum.
- Distance estimates to elliptical galaxies can achieve ~20% accuracy using this method.
- Large scatter in the traditional MMRD relation limits its usefulness as a standard candle.

## Abstract

Ten weeks of daily imaging of the giant elliptical galaxy M87 with the Hubble Space Telescope (HST) has yielded 41 nova light curves of unprecedented quality for extragalactic cataclysmic variables. We have recently used these light curves to demonstrate that the observational scatter in the so-called Maximum-Magnitude Rate of Decline (MMRD) relation for classical novae is so large as to render the nova-MMRD useless as a standard candle. Here we demonstrate that a modified Buscombe - de Vaucouleurs hypothesis, namely that novae with decline times t2 > 10 days converge to nearly the same absolute magnitude about two weeks after maximum light in a giant elliptical galaxy, is supported by our M87 nova data. For 13 novae with daily-sampled light curves, well determined times of maximum light in both the F606W and F814W filters, and decline times $t2 > 10 days we find that M87 novae display M(606W,15) = -6.37 +/- 0.46 and M(814W,15) = -6.11 +/- 0.43. If very fast novae with decline times t2 < 10 days are excluded, the distances to novae in elliptical galaxies with stellar binary populations similar to those of M87 should be determinable with 1 sigma accuracies of +/-20% with the above calibrations.