[Paper Review] Unveiling the nature of the unidentified gamma-ray sources

F. Massaro, R. D’Abrusco+5 more|arXiv (Cornell University)|Mar 13, 2013

Nuclear Physics and Applications4 citations

TL;DR

This paper presents a novel method to identify unidentified gamma-ray sources (UGSs) as likely gamma-ray blazars by leveraging their distinct infrared (IR) colors in the WISE 3D color space, known as the 'locus' or 'WISE Gamma-ray Strip.' The approach successfully associates 23 of 28 UGSs and many AGUs with blazar candidates, significantly advancing the identification of high-energy sources through IR-based classification, with strong consistency across statistical and multifrequency validation.

ABSTRACT

One of the main scientific objectives of the recent Fermi mission is investigating the origin of the unidentified gamma-ray sources (UGSs). Despite the large improvements of Fermi in the gamma-ray source localization with respect to the past gamma-ray missions, about 1/3 of the gamma-ray objects detected do not have yet an assigned counterpart a low energies. We recently developed a new association method to identify if there is a gamma-ray blazar candidate within the positional uncertainty region of a generic gamma-ray source. This method is entirely based on the discovery that blazars, the largest known class of gamma-ray sources, can be recognized and separated from other extragalactic sources on the basis of their infrared colors. Here we summarize the results obtained by applying our association procedure to the unidentified gamma-ray sources (UGSs) and to the active galaxies of uncertain type (AGUs) listed in the Second Fermi Large Area Telescope (LAT) catalog.

Motivation & Objective

• To resolve the long-standing challenge of identifying the nature of unidentified gamma-ray sources (UGSs), which constitute ~1/3 of Fermi-LAT detections.
• To overcome limitations of traditional association methods that rely on radio or X-ray counterparts, which are often missing or incomplete.
• To develop a robust, infrared-based classification method that leverages the unique IR color distribution of gamma-ray blazars.
• To validate the method by cross-referencing with multifrequency data and statistical models, ensuring high-confidence associations.
• To provide a systematic, reproducible approach for identifying blazar candidates in UGSs and active galaxies of uncertain type (AGUs).

Proposed method

• The method is based on the discovery that gamma-ray-emitting blazars occupy a distinct region in the 3D WISE infrared color space, termed the 'locus'.
• The 2D projection of this locus in the [3.4]-[4.6]-[12] μm color-color diagram is known as the 'WISE Gamma-ray Strip'.
• A new parametrization of the locus using principal component analysis was developed to improve detection accuracy using the full WISE all-sky survey data.
• The association procedure identifies potential blazar candidates within the positional uncertainty regions of UGSs and AGUs by checking if any source lies within the WISE color locus.
• The method is applied to the 2FGL catalog, including 28 UGSs and AGUs, with follow-up checks using archival data from radio, X-ray, and optical surveys.
• Statistical validation is performed by comparing results with Classification Tree and Logistic Regression models from Ackermann et al. (2012).

Experimental results

Research questions

• RQ1 Can infrared colors from the WISE survey reliably identify gamma-ray blazar candidates among unidentified gamma-ray sources?
• RQ2 How does the WISE-based method compare in performance and consistency with statistical association methods like Logistic Regression and Classification Trees?
• RQ3 To what extent can the WISE color locus reduce the ambiguity in classifying active galaxies of uncertain type (AGUs)?
• RQ4 What is the fraction of UGSs that can be confidently associated with blazar candidates using IR color criteria alone?
• RQ5 Are the identified WISE color-selected candidates consistent with known blazar characteristics across multiple wavelengths?

Key findings

• The method successfully associated 23 out of 28 unidentified gamma-ray sources (UGSs) with gamma-ray blazar candidates based on their WISE infrared colors.
• The association results show strong consistency with statistical models: 23 of the 28 WISE-identified candidates were also classified as active galaxies by Ackermann et al. (2012) with high confidence.
• A significant fraction of the candidates exhibit blazar-like features in the literature, including featureless or broad-lined optical spectra, flat radio spectra, and X-ray detections.
• The method’s success is attributed to the distinct clustering of gamma-ray blazars in the 3D WISE color space, which separates them from thermal-emitting sources.
• The approach remains robust even when traditional counterparts (radio, X-ray) are absent or ambiguous, improving identification rates over prior methods.
• The study highlights the need for future spectroscopic follow-up in optical and near-IR bands to confirm the nature of candidates and assess contamination from non-blazar sources.