Mariska Kriek

Associate Professor of Astronomy  |  University of California, Berkeley


FAST Download

  1. v1.0, Oct 28, 2013 (The,,, and libraries have to be downloaded separately).
    - FSPS models by Conroy et al. have been included. FSPS is only available for a Chabrier IMF, high resolution, and an exponential and delayed SFH, but the other models can be added by the user.
    - The Maraston models are now also available at high resolution for the exponentially declining SFH
    - The dust law is now a free parameter, you can choose between the Calzetti law, the Milky Way, the average dust law found by Kriek & Conroy (2013), and the free parameterization by Noll et al.
    - You can now also output the best-fit at spectrum the same resolution as the input data.
    - Several small bugs have been fixed. Two larger bugs are discussed in more detail below.
    - In v0.8 the following option was introduced: “If you have a small catalog and do not perform Monte Carlo simulations, FAST will replace your redshift grid by the input redshift array. This is faster, and allows more accurate redshifts.” However, this option previously only used photometric redshifts, which means, that in case of a spectroscopic redshift, the closest photometric redshift was assumed. This bug has been fixed, and spectroscopic redshifts are included as well. This bug did not affect you in case your catalog had more entries than the redshift grid, or you did not input a .zout, or you run monte carlo simulation.
    - The "MY_SFH" option should now work properly. Please read the instructions in the parameter file though for how to use this option

  2. v0.9b, May 2, 2011
    Fixed minor bug: only occurred when SAVE_CHI_GRID was set to 1 and z_phot or z_spec was not defined.

  3. v0.9, April 26, 2011
    Please note that column identifications that DO NOT contain fluxes cannot start with “F” or “E” and end with a number. If you have a column called “Fblabla[N]”, FAST will not work properly.

  4. v0.8d; December 16, 2010
    - Fixed bug: filter numbers ending on “0” were not read in
    - Due to quota issues, I had to remove and Please contact me if you would like to use these libraries, and we can find another way to transfer them.

  5. v0.8c; December 15, 2010
    Important revisions:
    - Fixed bug: In case TOT[N] was used, the “-99” fluxes in the input where not recognized as “no coverages”
    - There is more flexibility in input files (e.g., the translate files can have more than 2 columns, and FUV can be used as header name)

  6. v0.8b; August 26, 2010
    - Bug (introduced in v0.8) in determination of confidence intervals is fixed
    - Previously, the reduced chi^2 values did not take into account photometric bands with no coverage. This has been fixed in the current version

  7. v0.8; July 30, 2010
    Important revisions:
    - You can now choose the parametrization of the star formation history. Note that this has slightly changed the directory structure and parameter file
    - You can also input your own custom star formation history. You have to make it yourself using the galaxev software by Bruzual & Charlot (2003), and give the file the required name
    - If you have a small catalog, and do not perform Monte Carlo simulations, FAST will replace your redshift grid by the input redshift array. This is faster, and allows more accurate redshifts.
    - The confidence intervals on the input parameters (Av, age, tau, metallicity) are not cut off at real grid points, but determined by interpolation. This is slightly more accurate.

  8. v0.7; November 23, 2009
    Major revisions:
    - The filter file is updated (please check whether your filter numbers are still the same)
    - The “-1,-1,-1” bug (familiar to some of you ;-) has been fixed
    - There is the option of not imposing the age of the universe as the maximum age
    - You can define the column you want to read in as photometric redshift (default is now “z_phot”)

  9. v0.6; August 12, 2009
    - Major bug fixed: if FAST printed the following warning message: “WARNING: no solution for object xx”, your output catalogs are probably affected by this bug. If your input catalogs were in the right format, you normally won’t get this message, and this bug won’t have affected you.
    - It is now also possible to read in spectroscopic redshift from the [CATALOG].zout file
    - BEST_FIT directory is now placed in the OUTPUT_DIR
    - Small changes in the parameter file, directory structure, allowed label names etc.

  10. v0.5b; July 22, 2009

  11. v0.5; May 8, 2009: Only small changes made.

  12. v0.4; April 1, 2009: Header names are changed, and there is an option added to output the best fit.

  13. v0.3d; January 27, 2009: Several bugs fixed.

  14. v0.3c; January 9, 2009
    In the previous version the mass estimation for objects with no coverage in certain bands was wrong, when the template error function was used. This bug is fixed in this version. Furthermore, the filter file is updated.

  15. v0.3b; December 9, 2008
    Fixed bug in zeropoint parameter and in confidence level determination (when using EAzY). χ2 is also printed in the output file. Furthermore, it is now possible to determine more confidence intervals simultaneously!

  16. v0.3a; November 5, 2008
    The bug in the specific SFR has been fixed. Furthermore the directory structure has been slightly changed. On request of some of you, now the parameter file is also saved!

  17. v0.2d; October 7, 2008
    New feature: option to use the template error function.
    NOTE: bug in specific SFR and some broken links in spectrum example; new version will follow soon!

  18. v0.2c; August 12, 2008
    There are no major changes, but many small bugs are fixed.

  19. v0.2b; August 1, 2008
    The most significant changes (in addition to fixing many small and not so small bugs) are:
    - The master filter file is downsampled to R=300, and complemented by the UDS filters.
    - The ised.lib library contains all BC03, MA05 and CB07 models for all available metallicities and IMFs.
    - When fitting spectra, the continuum can be automatically scaled to the photometry.

  20. v0.2a; July 21, 2008
    The main improvements of this version are the following:
    - FAST is now an executable file.
    - The directory structure has changed.

  21. v0.1c; July 17, 2008
    In addition to a lot of minor changes, this version has the following new features:
    - FAST can be called from the command line; there is no need to start IDL first
    - All important information is in the parameter file, there is no README file anymore

  22. v0.1b; July 14, 2008
    Important changes compared to previous version:
    - Spectral fitting, optionally in combination with photometry, is fully working and tested
    - The output file contains grid and filter information, and is more read-in friendly
    - Updated master filter file (now similar as for EAzY)
    - The determination of the confidence intervals, when using EAzY has changed
    - There is the possibility of choosing 68%, 95% or 99% confidence intervals on the output parameters

  23. v0.1a; July 7, 2008

Home       CV       Publications       Research       Downloads