README file to accompany Worthey (1994) CD_ROM package. The package as described here should be regarded as copyright 1996 by Guy Worthey. Permission is given for free copying and alteration of the files here provided the intent is for scientific investigation. This collection of model grids is meant to allow users to explore the consequences to integrated light indices (colors, line strengths, M/L ratios, SFB magnitudes) as stellar population parameters (age mix, metal mix, IMF, helium abundance) are varied. Programs are provided that interpolate in the model grid and combine populations together. They are intended as a starting place for users. ---------------------------------- 1. CHANGES FROM WORTHEY (1994) For this CD-ROM, only two changes have been made to the original Worthey (1994, =W94) models. First, for metal-poor populations, the morphology of the horizontal branch now follows the behavior of the Galactic globulars, becoming extended toward the blue. M3 and M92 were used as models for this behavior as in Aaronson et al. (1978), the M3 horizontal branch extending to log Te=4, and the M92 horizontal branch extending from log Te=3.9 to log Te=4.1. The switch to M3 morphology happens below [Fe/H] = -0.95; the switch to M92 morphology happens below [Fe/H] = -1.55. This scheme is discussed in Worthey (1993). Second, the $M/L$ ratios have been put on a less arbitrary scale. The lower mass-cutoff of the power-law IMFs was varied to make the models match a globular cluster M/L_V = 2.5. The upper mass cutoff was assumed to be 10 Mo. It is also assumed that there are no stellar remnants above this mass. Below 10 Mo the Weidemann (1987) initial-to-final mass relation was assumed. Here are the lower mass-cutoffs as a function of power-law IMF slope: Power-law slope lower mass cutoff 1.35 0.33 Mo 1.85 0.18 2.35 (Salpeter) 0.21 2.85 0.28 3.35 0.33 The normalization of the mass follows W94: the INITIAL mass of the system, obtained by integrating under the IMF, is normalized to be a million Mo. The mass of the population changes somewhat as stars are converted into remnants, and it is this reduced mass which is used to calculate the M/L ratios. There is a paper in preparation which focusses on the M/L issue. These two changes were wrought because of their obvious utility (for instance, the M/L ratios may now have some physical meaning!). The temptation to change underlying isochrones or fluxes was resisted in order to have all of the details in W94 apply to this CD-ROM distribution. One last detail: the new definitions for H gamma and H delta are included in the model output. There is some risk that these model predictions will change a little bit in the future because the paper which describes the stellar data and fitting functions is not yet refereed (Worthey & Ottaviani 1996). ---------------------------------- 2. THE WEB If you are only an occasional consumer of population models, the web page http://www.astro.lsa.umich.edu/users/worthey/getmodels/dial_a_model.html may be all you need: it is a working version of the software on this CD-ROM connected to the display mechanism of HTML browsers. ---------------------------------- 3. THE DIRECTORY TREE AND WHAT IS THERE The directory on the CDROM that this README file is in I am calling the "root directory." It contains a few fortran modules, a makefile, and 11 subdirectories. One subdirectory contains William Pence's FITSIO routines. The rest contain collections of FITS format files, each of which has data for a population model of one age and one metallicity. The file names are formed from the age and Z. The FITS files are composed of several batches of header information, a binary array of model fluxes, and 3 ascii table extensions with index data, color/SBF data, and the underlying color-magnitude diagram. THERE IS MORE INFORMATION IN THE FITS FILES THAN IS EXPLICITLY READ BY THE FORTRAN PROGRAMS. Users are encouraged to browse through a FITS file and modify the programs as they see fit. The fortran programs for interpolation should be set up as described below. The Wo_FITSIO directory has its own README, fortran routines, and a makefile geared for unix users. A branch directory "Pe_FITSIO" contains the original FITSIO distribution, the user manual, VMS-specific code, and slick makefiles by Steve Allen. We have: Root directory: Wo_-README | this README file Wo_ask.f | driver program Wo_getmodel.f | fortran module Wo_locate.f | fortran module Wo_rfits.f | fortran module Wo_makefile | makefile for unix systems Wo_FITSIO/ dir FITSIO modules, makefile, Pe_FITSIO subdir Wo__vanilla/ dir Zero models. Salpeter IMF, dY/dZ = 2.7 Wo_MillerSc/ dir Miller-Scalo (1979) IMF Wo_Salp+0.5/ dir IMF slope up 0.5. slope = 2.85 Wo_Salp+1.0/ dir IMF slope up 1.0. slope = 3.35 Wo_Salp-0.5/ dir IMF slope down 0.5. slope = 1.85 Wo_Salp-1.0/ dir IMF slope down 1.0. slope = 1.35 Wo_Ymaximum/ dir High helium content. Wo_Yminimum/ dir Low helium. Wo_Yshallow/ dir dY/dZ=6 at low Z, dY/dZ=0 at high Z. Wo_NoLateSt/ dir No horizontal branch or later stages Wo_original/ dir W94 exactly as tabulated ---------------------------------- 4. EXPLANATION OF MODEL FLAVORS (a) Wo__vanilla, Wo_original These two directories differ from each other as described above: "original" has no provision for stellar remnants, its IMF is integrated from 0.1 to 2.0 solar masses, and all of this mass is assumed to contribute to the M/L ratios. The horizontal branch is assumed to lie in a clump at all metallicities in the "original." "Vanilla" has a Salpeter IMF integrated from 0.21 to 10.0 solar masses, stellar remnants, and a horizontal branch that varies as a function of metallicity. (b) Wo_Salp+0.5, Wo_Salp+1.0, Wo_Salp-0.5, Wo_Salp-1.0, Wo_MillerSc These directories have variations of IMF. The lower mass cutoff is 0.1 Mo for the Miller & Scalo (1979) IMF, otherwise is given by the table in section 1. The upper mass cutoff is 10 Mo. (c) Wo_Yminimum, Wo_Ymaximum, Wo_Yshallow Helium abundance is a free parameter in the W94 models. It is allowed to affect stellar lifetimes and temperatures through the helium flash. It affects later stage lifetimes through modulation of the helium core mass, but any temperature effects are neglected for later stages. The Y_minimum and Y_maximum flavors are set by the helium abundances chosen by VandenBerg in the input evolutionary isochrones. He chose Y=0.20 and Y=0.30 below Z=0.008, and Y=0.25 and Y=0.35 above that value. (In the near-solar-Z regime some interpolation was needed. See W94.) The Y_shallow scheme is that Y = 0.228 + 6*Z below Z=0.0077 and Y = 0.274 above that value. The Z=0.0077 corresponds to [Fe/H] = -0.34. (VandenBerg solar Z = 0.0169. See W94.) The default scheme in "vanilla" is Y = 0.228 + 2.7*Z at all Z. (d) Wo_NoLateSt All stages of stellar evolution beyond the helium flash are truncated. ---------------------------------- 5. INTERPOLATION SOFTWARE The fortran programs driven by Wo_ask.f will interpolate a model of any [Fe/H] and age from any of the available model flavors. The program is interactive and will walk the user through the process. Users may discard the "ask" driver and use the rest of the subroutines directly if they so desire. The makefiles were designed for unix platforms, but the fortran is standard, and should run on any platform. (Now that I think of it, there is one exception: a routine in Wo_ask.f called ctime is unix-specific, and maybe even Sun-specific, but all it does is return the time and date, so it can easily be deleted or replaced.) To get the interpolation programs working: 1. Copy all *.f programs, the Wo_makefile file, and the entire Wo_FITSIO directory to your own disk space. These instructions assume that you have left the directory structure and file names intact. 2. rename "Wo_makefile" to "makefile." If you have a VMS machine, the makefile will not work - you are on your own. (I have little experience with VMS. The programs were debugged and tested on Sun workstations running SUNOS and SOLARIS.) 3. rename "Wo_FITSIO/Wo_Makefile" to "Makefile." Type "make." This should make an object library called "libfitsio.a." The code for VMS users is in the next directory: "Pe_FITSIO." 4. IMPORTANT: Edit Wo_ask.f, and change the data for the variable "rootdir" to reflect the absolute pathname of the directory where the Wo_ask.f file sits on the cdrom. Also edit the variable "nchar" that keeps track of the number of characters in the "rootdir" string. 5. Do the final compile with a "make" in the Wo_ask.f directory. Invoke the program with "Wo_ask." ------------------------------ 6. ACKNOWLEDGEMENTS I would like to thank Claus Leitherer, Uta Fritze-von Alvensleben, and John Huchra, first for a good conference, and second for providing the impetus for me to provide these grids, a task which I always meant to do, but never made a first priority. Thanks also to Philip Hughes, and, historically, to Steve Allen. They both provided me with needed files and stellar technical advice. ------------------------------ 7. REFERENCES Aaronson, M., Cohen, J. G., Mould, J. R., Malkan, M. 1978, ApJ, 223, 824 Miller, G. E., & Scalo J. M. 1979, ApJS, 41, 573 Worthey, G. 1993, ApJL, 415, L91 Worthey, G. 1994, ApJS, 95, 107; W94 ------------------------------ 8. SECTION FROM THE LEITHERER ET AL. PASP PAPER Worthey (1994, ApJS, 95, 107) models The Worthey (1994) evolutionary population models give integrated UBVR$_C$I$_C$JHKLL$'$M colors and magnitudes, SBF magnitudes, spectral energy distributions, and Lick/IDS spectral index strengths for populations of arbitrary age, and (assumed scaled-solar) abundance [Fe/H]. The ages range from 1 to 18 Gyr, and abundances from -2.0 to +0.5 dex. Helium abundance $Y$ is a free parameter. The underlying stellar evolution is that of VandenBerg (VandenBerg 1985, VandenBerg \& Bell 1985, VandenBerg \& Laskarides 1987). The stellar fluxes used to compute integrated flux are mostly theoretical (Kurucz 1992, Bessell et al. 1989, 1991), and explicitly include the effects of metallicity on the spectral shape and colors. Please see Worthey (1994) for details. The distribution on the CD-ROM includes several ``flavors'' of model grid: (1) Five power-law IMFs plus Miller \& Scalo (1979) are provided. For the power-law IMFs the lower mass cutoff was chosen so that $M/L_V=2.5$ for globular cluster-like populations (with a reasonable accounting of stellar remnants; Worthey 1996, in preparation), so output $M/L$ values differ from the tabulation in Worthey (1994) by a constant factor plus a small perturbation caused by the fact that a small amount of low-mass starlight is missing. (2) Also included are four schemes for how helium abundance tracks overall abundance. (3) One model grid without helium-burning and later stages of stellar evolution is included for those interested in adding HB-, AGB-, EAGB-, PAGB-stages, or other stages of evolution by themselves. (4) The models as tabulated in Worthey (1994) are also included unaltered. Each model (at one age and [Fe/H]) is stored in a FITS file with ascii table extensions. One ``flavor'' is made of a grid of FITS files covering the range of age and [Fe/H]. A fortran program is provided that reads the data using William Pence's FITSIO routines (also included), interpolates in the model grid to arbitrary age and metallicity, and combines populations in any combination the user specifies. Multicolor color-magnitude diagrams are also available in the FITS file ascii extensions. If the reader is but an occasional consumer of population models, he or she would probably best be served by the web page http://www.astro.lsa.umich.edu/users/worthey/getmodels/dial_a_model.html which contains a form-driven version of the interpolation/addition program on this CD-ROM. Bessell, M. S., Brett, J. M., Scholz, M., \& Wood, P. R. 1989, A&AS, 77, 1 Bessell, M. S., Brett, J. M., Scholz, M., \& Wood, P. R. 1991, A&AS, 89, 335 Kurucz, R. L. 1992, private communication Miller, G. E., \& Scalo J. M. 1979, ApJS, 41, 573 VandenBerg, D. A. 1985, ApJS, 58, 561 VandenBerg, D. A., \& Bell, R. A. 1985, ApJS, 58, 561 VandenBerg, D. A., \& Laskarides, P. G. 1987, ApJS, 64, 103 Worthey, G. 1994, ApJS, 95, 107