We present a new technique for accurately determining the ages of starbursts in post-starburst galaxies. In particular, it uses the strength of the Ca II H+Hη absorption feature relative to that of Ca II K to separate the effects of burst strength from burst age in a spectrum comprised of a post-starburst population and an underlying old galaxy population. The technique is based on comparing the integrated spectra of post-starburst galaxies with starburst model spectra produced from the evolutionary synthesis code of Bruzual & Charlot (1995, ApJ, in preparation). Model spectra have been generated for starbursts viewed at a variety of times (ranging from 0.0 to 2.0 Gyr) after completion of the burst. Each of these spectra has been combined in various amounts with observed integrated spectra of elliptical galaxies to simulate different combinations of starbursts and underlying old populations. To test the technique we have measured spectral indices in the observed spectra of three post-starburst early-type galaxies. The three galaxies are shown to have different post-starburst ages, ranging from 0 to 1.5 Gyr after completion of the burst. We also demonstrate that if the burst contributes greater than 50% of the light at 4000 A then the technique can distinguish between a post-starburst system and models in which a constant star formation history is suddenly truncated, such as that which could occur in a spiral that is rapidly stripped of its interstellar medium. Finally, we demonstrate that while the technique is well suited for determining the time elapsed since the termination of a starburst (up until 2 Gyr after the end of the burst), the derived termination times are relatively insensitive to the duration of the burst.
No takes yet. Share an insight, caveat, or question.
A 1996 study studied this question.