ИСТИНА |
Войти в систему Регистрация |
|
ИСТИНА ЦЭМИ РАН |
||
We present the new technique for the determination of the low-mass slope of the stellar mass function using the pixel-space fitting of integrated light spectra. This method can be used to constrain the stellar initial mass function (IMF) of compact stellar systems having high relaxation timescales (Hubble time or more). The technique can be used in two variants. In the first case the best-fitting template is represented by a simple stellar population (SSP) with free age, metallicity, and the IMF, and all parameters are returned in a single minimization loop, using the pre-computed grid of the high-resolution PEGASE.HR SSP models for Kroupa canonical IMF with variable alpha. The second way exploits the independently determined mass-to-light ratio of the stellar population (i.e. the dynamical estimates for dark matter free systems like globular clusters or ultra-compact dwarf galaxies) which we use to form a two dimensional grid of models for a set of ages and metallicities having this pre-defined M/L. This grid is then supplied to the standard NBursts full spectral fitting technique. We used Monte-Carlo simulations with mock spectra to test both approaches and conclude that: (1) age, metallicity, and IMF can be very precisely determined in the first ``unrestricted'' variant of the code for high signal-to-noise datasets (S/N=100, R=7000 give the uncertainty of alpha of about 0.1); (2) adding the M/L information significantly improves the precision and reduces the degeneracies, however systematic errors in M/L will translate into offsets in alpha. We also applied the technique to observed intermediate and high-resolution GG and UCD spectra and demonstrated its applicability to real data. We conclude that by applying our technique to high-quality optical observations of extragalactic GCs and UCDs we are able to reach better precision of the IMF determination than that made with direct star counts in nearby open clusters and check the IMF universality hypothesis.