Contents
This page describes the Fuji
value-added catalog which was publicly released
in December 2023 and is based on the DESI Early Data Release (DESI/EDR).
Please refer to the :ref:`acknowledgments` section for the conditions for using this VAC.
Note
We only document the latest version of this VAC here; please see below for information regarding :ref:`previous versions<previous versions - fuji>`.
Data from the Fuji
VAC can be accessed at any of the following links:
Data url | https://data.desi.lbl.gov/public/edr/vac/edr/fastspecfit/fuji/v3.1 |
Web-app (for collaborators) | https://fastspecfit.desi.lbl.gov |
NERSC (for collaborators) | /global/cfs/cdirs/desi/public/edr/vac/edr/fastspecfit/fuji/v3.1 |
For more information regarding the content and organization of the VAC, please click on the following links:
- :ref:`Healpix Catalogs<healpix catalogs>`
- :ref:`Merged Catalogs<merged catalogs>`
- :ref:`Sample Selection<sample selection>`
- :ref:`Updated QSO Redshifts<qso redshifts>`
The following table summarizes the size and number of targets in each merged catalog.
Note
The last catalog listed in the table is a super-merge of all the preceding catalogs, i.e., a merge over all surveys and programs.
.. rst-class:: columns
File Name | File Size | Number of Targets |
---|---|---|
fastspec-fuji-cmx-other.fits | 12.8 MB | 2,771 |
fastspec-fuji-special-dark.fits | 161 MB | 35,649 |
fastspec-fuji-sv1-backup.fits | 17.1 MB | 3,683 |
fastspec-fuji-sv1-bright.fits | 579 MB | 126,678 |
fastspec-fuji-sv1-dark.fits | 1.05 GB | 235,942 |
fastspec-fuji-sv1-other.fits | 156 MB | 34,152 |
fastspec-fuji-sv2-backup.fits | 664 KB | 107 |
fastspec-fuji-sv2-bright.fits | 213 MB | 46,510 |
fastspec-fuji-sv2-dark.fits | 241 MB | 52,781 |
fastspec-fuji-sv3-backup.fits | 7.31 MB | 1,564 |
fastspec-fuji-sv3-bright.fits | 1.19 GB | 265,325 |
fastspec-fuji-sv3-dark.fits | 2.66 GB | 592,441 |
fastspec-fuji.fits | 6.29 GB | 1,397,603 |
The following table summarizes the number of QSO targets whose redshift has been updated using the procedure documented :ref:`here<qso redshifts>`.
.. rst-class:: columns
Catalog | Number of Objects | Number with Corrected Redshifts |
---|---|---|
{fastspec}-fuji-cmx-other.fits | 2,771 | 56 |
{fastspec}-fuji-special-dark.fits | 35,649 | 313 |
{fastspec}-fuji-sv1-backup.fits | 3,683 | 100 |
{fastspec}-fuji-sv1-bright.fits | 126,678 | 65 |
{fastspec}-fuji-sv1-dark.fits | 235,942 | 3,810 |
{fastspec}-fuji-sv1-other.fits | 34,152 | 170 |
{fastspec}-fuji-sv2-backup.fits | 107 | 0 |
{fastspec}-fuji-sv2-bright.fits | 46,510 | 8 |
{fastspec}-fuji-sv2-dark.fits | 52,781 | 1,029 |
{fastspec}-fuji-sv3-backup.fits | 1,564 | 0 |
{fastspec}-fuji-sv3-bright.fits | 265,325 | 133 |
{fastspec}-fuji-sv3-dark.fits | 592,441 | 3,444 |
{fastspec}-fuji.fits | 1,397,603 | 9,128 |
The following tables document the code versions and environment variables used
to produce this VAC. For details regarding the revision history of
FastSpecFit
, please see the change log.
Note that the tagged dependencies can be retrieve from any FITS file with the following bit of code:
import fitsio from desiutil.depend import Dependencies codever = Dependencies(fitsio.read_header('/path/to/fastspecfit/file.fits, ext=0)) for codename, version in codever.items(): print(codename, version)
.. rst-class:: columns
Software Package | Version(s) |
---|---|
python | 3.10.8 |
numpy | 1.22.4 |
scipy | 1.8.1 |
astropy | 5.2.1 |
yaml | 6.0 |
matplotlib | 3.6.2 |
fitsio | 1.1.8 |
desiutil | 3.4.2 |
desispec | 0.60.2 |
desitarget | 2.6.1 |
desimodel | 0.19.0 |
speclite | 0.17 |
fastspecfit | 2.4.3 |
.. rst-class:: columns
Environment Variable | Value |
---|---|
DESI_ROOT | /dvs_ro/cfs/cdirs/desi |
DUST_DIR | /dvs_ro/cfs/cdirs/cosmo/data/dust/v0_1 |
FPHOTO_DIR | /dvs_ro/cfs/cdirs/desi/external/legacysurvey/dr9 |
FTEMPLATES_DIR | /dvs_ro/cfs/cdirs/desi/science/gqp/templates/fastspecfit |
FTEMPLATES_FILE | ftemplates-chabrier-1.2.0.fits (see README.txt) |
FPHOTO_FILE | /global/common/software/desi/perlmutter/desiconda/20230111-2.1.0/code/fastspecfit/2.4.3/lib/python3.10/site-packages/fastspecfit/data/legacysurvey-dr9.yaml |
EMLINES_FILE | /global/common/software/desi/perlmutter/desiconda/20230111-2.1.0/code/fastspecfit/2.4.3/lib/python3.10/site-packages/fastspecfit/data/emlines.ecsv |
In this section we briefly document issues or problems which were identified with the VAC after its final release, and provide additional details regarding previous versions of this VAC.
Please see the change log for a record of what code and data model changes have occurred since these versions were released and to report additional problems or to request new features please open a ticket.
- Release date: December 2023
FastSpecFit
version:2.4.3
- Templates:
ftemplates-chabrier-1.2.0.fits
(see README.txt). - Notes:
- Minor updates to spectrophotometric templates (expanded velocity dispersion grid; see PR/#158).
- Known Issues:
- Warning: Stellar masses are systematically higher (by 0.2-0.5 dex) compared to other methods, so they should be used with caution (see issue/#159). Similarly, star-formation rates and other SPS model parameters have not been fully validated.
- Release date: August 2023
FastSpecFit
versions:2.4.1
,2.4.2
- Templates:
ftemplates-chabrier-1.1.0.fits
(see README.txt). - Notes:
- Minor updates to spectrophotometric templates.
- Just two rounds of emission-line fitting, not three (see PR/#151).
- Updated IGM attenuation coefficients (see PR/#136).
- Major algorithmic updates related to how emission-line amplitudes, fluxes, and inverse variances are computed, including a bug fix which the emission-line model spectra were not being convolved with the resolution matrix (see PR/#137).
- Known Issues:
- Warning: Stellar masses are systematically higher (by 0.2-0.5 dex) compared to other methods, so they should be used with caution (see issue/#159). Similarly, star-formation rates and other SPS model parameters have not been fully validated.
- Bug: Fluxes (and EWs) of lines which lie in the camera-overlap region are overestimated by a factor of 2 due to a bug handling the different pixel scale (fixed in PR/#157).
- Release date: February 2023
FastSpecFit
versions:2.1.0
,2.1.1
- Templates:
ftemplates-chabrier-1.0.0.fits
(see README.txt). - Notes:
- Major update of underlying spectrophotometric templates.
- Known Issues:
- Release date: August 2022
FastSpecFit
versions:1.0.0
,1.0.1
- Templates: SSP-CKC14z/v1.0
- Notes & Known Issues:
- First major release; many issues addressed in future releases.