We study how the spectral fitting of galaxies, in terms of light fractions derived in one spectral region translates into another region, by using results from evolutionary synthesis models. In particular, we examine propagation dependencies on Evolutionary Population Synthesis models, age, metallicity, and stellar evolution tracks over the near-UV to near infrared spectral region. Our main results are: as expected, young stellar population fractions derived in the optical cannot be directly compared to those derived in the NIR, and vice versa. In contrast, intermediate to old age fractions are similar over the whole spectral region studied. Spectral fitting performed in a restricted spectral range should not be directly propagated from the NIR to the UV/Optical, or vice versa. We provide equations and an on-line form (Panchromatic Averaged Stellar Population - PaASP) to be used for this purpose. The PaASP form is avaliable here.

Further details: Panchromatic Averaged Stellar Populations - Riffel et al. (2001)


This cl script is intended to reduce data from the OSIRIS spectrometer, taken in cross-dispersed mode. The code needs the science exposures in an arbitrary nodding pattern, a normalized flat image, a bias exposure and star exposures, for telluric line removal, also in a nodding pattern. Wavelength calibration is achieved via fitting of a dispersion function to OH lines present in the sky spectra, eliminating the necessecity for a separate lamp exposure. The result is a group of 1D spectra, one for each of the three apertures that roughly represent the J,H and K bands. Up to now the reference for XD-Spres is the paper: Near-infrared integrated spectra of Galactic globular clusters: testing simple stellar population models. Details are presented in XDSPRES: CL-based package for Reducing OSIRIS Cross-dispersed Spectra.

XD-Spres can be downloaded here.


We present Perl Algorithm to Compute continuum and Equivalent Widths (PACCE). We describe the methods used in the computations and the requirements for its usage. We compare the measurements made with PACCE and "manual" ones made using IRAF splot task. These tests show that for SSP models the equivalent widths strengths are very similar for both measurements. In real stellar spectra, the correlation between both values is still very good, but with differences of up to 0.5A. PACCE is also able to determine mean continuum and continuum at line center values, which are helpful in stellar population studies. In addition, it is also able to compute the uncertainties in the equivalent widths using photon statistics.

We ask that users of PACCE please reference the Riffel & Vale (2011) paper in any publications. Since this is the only recognition we receive for the time and effort we have spent developing it.

PACCE is fully described in: Riffel & Vale (2011).

Instaling PACCE:

The code can be instaled to be useded by any user by coping it as root to /usr/local/bin. However, it can be used as a normal user under his/her account.
Note: Do not forguet to make it an executable ($ chmod +x

Running PACCE:

PACCE can be download here as well as an input table example.

  • Perl
  • Perl's "Math::Derivative" package;
  • Perl's "Math::Spline" package;
  • Gnuplot;

  • All the requirementes are avaliable in Synaptic under DEB linux systems.

    To run pacce type in the shell:

    $ perl pacce --> if you are running it into your home directory.

    $ ./pacce --> if you make it an executable file and you are running it into your home directory.

    $ pacce --> if it is installed in the /usr/local/bin path (recommended).

  • Due to perl limitation our code produces a illegal division by zero if you are using fluxes in units of ~10^-15. We are working in a internal routine to normalize the spectra. Up to we do this, we recomend you to multiply the fluxes of your spectra by 10^15 (e.g awk '{print $1 " "$2*1.0e15}' < inpSpec > outSpec)
  • We reccomend to put your spectra into 1A bins.