Skip to main content
FSSC: Data : Data Access : LAT Third Catalog

LAT 4-year Source Catalog (3FGL)

The Large Area Telescope (LAT) on board NASA's Fermi Gamma-ray Space Telescope (launched June 11, 2008) observes in the 20 MeV to 300 GeV energy range. This web page presents the third full catalog of LAT sources, based on the first 48 months of survey data. For a full explanation about the catalog and its construction see the LAT 4-year Catalog Paper (3FGL) available on arxiv.

The source designation is 3FGL JHHMM.m+DDMM(c,e,i,s) where the 3 refers to the third catalog (1FGL was released at 1 year, and 2FGL at 2 years). Another is planned for 6+ years of data) and FGL represents Fermi Gamma-ray LAT. The optional "c", "e", "i" and "s" designators are explained in the caveats below.

LAT Catalog Data Products

The LAT 4-year Source Catalog is currently available as a FITS file and as an XML model file to be used for data analysis within the Fermi Science Tools. A BROWSE table containing a searchable version of the catalog will be made available soon. Supporting tools and documentation have been provided and are linked below.

It is important that all users of this catalog review the caveats listed below the catalog links. These describe the content of the catalog at a high level, as well as some cautions for the user.

We emphasize several important points and caveats regarding use of the 3FGL catalog:

  • This work is a summary of the LAT results covering the time period August 4, 2008, to July 31, 2012.
  • The analysis is based on data in the 100 MeV to 300 GeV energy range.
  • Source detection is based on the integrated data set, i.e., sources are detected according to their average fluxes.
  • The catalog adopted multiple spectral shapes, depending on the observed source properties. Pulsars are modeled by an exponentially cutoff Power Law. Others are modeled as a LogParabola if statistically significant curvature (according to Signif_Curve) is detected, or as a simple Power Law otherwise.
  • After computing the maximum likelihood, the threshold for inclusion is likelihood Test Statistic TS > 25, corresponding to somewhat more than 4 sigma significance. The catalog has 3033 sources.
  • The corresponding flux limit varies widely according to the hardness of the source spectrum and the brightness of the background Galactic diffuse emission.
  • The 3FGL information in five energy bands is not appropriate for detailed spectral modeling. Each entry represents the bast fit flux value, with corresponding positive (upper) and negative (lower) 1σ errors. In cases where the lower limit would go below zero, that value is NULL in the file. Note that some sources have only upper limits in each of the 5 bands; these sources have been detected based on their integrated fluxes.
  • The variability index indicates that the source is variable on a time scale of months. It does not address shorter or longer time variations. An index > 72.44 indicates a >99% confidence probability that the source is variable.
  • The diffuse model used in the analysis is still subject to improvements. Matching the model to the large-scale emission is an iterative process. The diffuse model is particularly important for sources from low to mid Galactic latitudes.
  • Catalog entries include two important quality flags that indicate potential issues: a "c" following the 3FGL name indicates that the source is found in a region with bright and/or possibly incorrectly modeled diffuse emission; any non-zero entry in the "Flags" column indicates that some inconsistency was seen during the analysis (see the table in the preprint for details). Sources with either of these indicators should be used with great care. They correspond to significant excesses of photons, but such excesses can result from residual extended emission or confused source pile-up.
  • The catalog analysis assumed point source emission for all but twenty-five extended sources, which were modeled using spatial templates and are indicated by an "e" following the name. Results for extended sources such as the LMC and Cen A are likely to not be representative of the total emission, and additional catalog point sources may appear near such objects.
  • The Crab pulsar and PWN are represented by a total of three entries, two of which (designated "i" for inverse compton and "s" for synchrotron) represent spectral components of the PWN. We consider these three entries to represent two sources.
  • Associations listed in the catalog are not in general to be taken as firm identifications. Associations are positional coincidences that are statistically unlikely to be due to chance alignment between known or candidate gamma-ray producing objects with 3FGL sources. Except for those cases with correlated variability or spatial extent, a physical relationship is not established. The designators for some associated sources may be updated in a future release to use more common or more complete names.

Again, for more information about the catalog see the LAT 4-year Source Catalog Paper. Feedback about the catalog and paper should be directed to the paper contact authors. Ongoing efforts to understand the calibration and improve the analysis techniques may lead to updates to the 3FGL catalog. Any significant updates will be advertised to the fermi-news mailing list. All changes will be documented in the change log below.

LAT 4-year Source Catalog (3FGL) Change Log

Rev Date Filename Change Description
1 09 Jan 2015 gll_psc_v14.fit Initial Release
2 30 Apr 2015 gll_psc_v15.fit Updated several associations, released full content of Table 11
3 18 May 2015 gll_psc_v16.fit Added TUCDnn keywords to the main FITS file.