Constraining the gamma-ray efficiency of LINER outflows with Fermi-LAT and MEGARA

This paper investigates whether the extended ionized gas outflows in Low-Ionization Nuclear Emission-line Regions (LINERs) can efficiently accelerate particles to relativistic energies and power high-energy gamma-ray emission, or whether a compact nuclear jet is required. The authors combine spatially resolved optical integral-field kinematics from MEGARA at the Gran Telescopio Canarias, which provide kinetic powers of ionized outflows (ĖOF), with 17 years of Fermi-LAT observations to derive 0.05–500 GeV luminosities or 95% confidence upper limits for a sample of six local LINERs.

The sample comprises five radio-quiet to moderately radio-emitting LINERs (NGC 3226, NGC 3245, NGC 4278, NGC 4438, and NGC 4750) and one radio-loud LINER (NGC 1052), which hosts a prominent sub-parsec twin jet and serves as an internal control. The kinetic powers of the ionized outflows are adopted from Hermosa Muñoz et al. (2024) and Cazzoli et al. (2022), and are lower limits to the total mechanical energy available for particle acceleration.

The Fermi-LAT analysis of 17 years of Pass 8 data yields no significant detections (T S ≥ 16) for any of the six LINERs, although three sources (NGC 3226, NGC 4278, and NGC 4750) show marginal hints of emission (9 < T S < 16). The authors therefore compute 95% confidence level upper limits on the 0.05–500 GeV luminosities for the entire sample. The most physically constraining limit is found for the radio-loud LINER NGC 1052, where the maximum efficiency is restricted to η < 41%. For the rest of the sample, the Fermi-LAT upper limits generally lie well above the kinetic power of the ionized outflows (η ≫ 100%).

The authors construct a diagnostic diagram comparing Lγ and kinetic power, placing their sample in the context of archetypal starbursts (M82, NGC 253) and radio galaxies (Centaurus A, M87). The starbursts populate the low-efficiency region (η ≲ 1%), with M82 showing η ≈ 0.7%. For most of the radio-quiet LINERs, the upper limits sit well above the measured ionized outflow powers, and at starburst-like efficiencies their expected luminosities would fall three to four orders of magnitude below current GeV thresholds. The non-detections are therefore consistent with extended LINER outflows acting as standard, low-efficiency accelerators.

The paper highlights the recent very-high-energy γ-ray detection of NGC 4278, which showed extreme radiative efficiencies that challenged standard shock-driven emission models. The authors note that the efficiency inferred for NGC 4278 (η ≲ 2 × 104%) cannot be sustained by its ionized outflow, even allowing for its kinetic power being a lower limit. The segregation of NGC 4278, M87, and Cen A from the outflow-dominated region of the diagram indicates that efficient γ-ray emission requires a compact nuclear jet. M87 is particularly instructive, since it is itself classified as a LINER: its position in the diagram is set by the large power channeled through its jet, while its efficiency (η ≈ 2.6%) remains comparable to that of the starbursts. It is thus the power available to a compact jet, rather than an anomalous efficiency, that makes a LINER detectable at GeV energies.

The authors conclude that the extended ionized outflows in LINERs are highly inefficient high-energy particle accelerators, analogous to starburst superwinds. These sample-level constraints demonstrate that extreme putative efficiencies cannot be sustained by the ionized outflows alone, favoring a compact nuclear jet origin for the most efficient γ-ray emitting LINERs. They note that at starburst-like efficiencies (η ∼ 0.7%), an outflow would need ĖOF ≳ 1.4 × 1041 erg s−1 to reach a detectable γ-ray luminosity of ∼ 1039 erg s−1, and future observations with the Cherenkov Telescope Array Observatory (CTAO) will be crucial to map the boundary between outflow-driven and jet-driven feedback at the faint end of the AGN population.

Improvements to AI Systems:

1. Astrophysical Source Classification and Efficiency Prediction

• Train a multi-label classifier on multi-wavelength data (radio, optical IFU kinematics, gamma-ray) to automatically distinguish between outflow-dominated and jet-dominated LINERs.

• The improved system can predict whether a given LINER’s gamma-ray emission is likely powered by extended outflows (low efficiency, η 100%), reducing false positives in source identification.

2. Bayesian Upper-Limit Inference for Non-Detections

• Implement a hierarchical Bayesian model that jointly fits Fermi-LAT photon counts, kinetic power measurements (with uncertainties), and efficiency priors to compute posterior upper limits on gamma-ray luminosity for non-detected sources.

• The system can output robust 95% confidence intervals for η even when TS < 16, enabling statistically rigorous comparisons across heterogeneous samples without requiring detections.

3. Automated Diagnostic Diagram Generation and Outlier Detection

• Build a tool that ingests Lγ, kinetic power, and jet power estimates for any AGN sample, then automatically generates the Lγ–ĖOF diagnostic diagram with efficiency contours (η = 0.1%, 1%, 10%, 100%).

• It can flag outliers (e.g., NGC 4278) that deviate from the outflow-driven scaling relation, prompting targeted follow-up for jet-dominated candidates.

4. Physical Consistency Checker for Gamma-Ray Emission Models

• Develop a rule-based reasoning module that cross-checks claimed gamma-ray detections against available kinetic power limits, applying the constraint η = Lγ/ĖOF < 100% (or a user-defined threshold).

• The system can automatically reject or flag astrophysical models that require impossible efficiencies (e.g., η > 104%) for a given outflow power, as demonstrated for NGC 4278.

5. Sensitivity Forecasting for Future Observatories (CTAO)

• Create a simulation tool that, given a source’s ĖOF and assumed efficiency (e.g., 0.7%), predicts the required integration time and detectability with CTAO, based on its energy-dependent effective area and background.

• The improved system can prioritize targets for CTAO observations by ranking LINERs where outflow-driven gamma-ray emission would become detectable, optimizing telescope time allocation.

6. Multi-Messenger Joint Fitting with Uncertainties

• Integrate optical IFU kinematic maps (e.g., MEGARA data) with Fermi-LAT likelihood profiles in a joint Bayesian framework, treating kinetic power as a lower limit (censored data) and gamma-ray flux as upper limits.

• The system can produce posterior distributions for the true mechanical energy and acceleration efficiency, propagating systematic errors from extinction, electron density, and outflow geometry.

7. Template-Based Classification of AGN Feedback Mechanisms

• Use the paper’s starburst (η 0.7%) and jet (η 2.6% but high power) archetypes to train a support vector machine or random forest on features like radio loudness, jet extent, and outflow velocity dispersion.

• The system can automatically classify new LINERs into “outflow-dominated,” “jet-dominated,” or “intermediate” categories, aiding large surveys like SDSS-V or WEAVE.

8. Automated Literature Synthesis for Efficiency Constraints

• Build a natural language processing pipeline that extracts ĖOF, Lγ, and jet power values from published papers (including this one) and updates a live database of AGN acceleration efficiencies.

• The system can then generate real-time meta-analyses, identifying which source classes consistently violate physical efficiency bounds, and suggest new theoretical models or observational strategies.