Fermi Large Area Telescope Detection of Gamma Rays from the NGC 6251 Radio Lobe

We report the detection of extended γ-ray emission from lobes in the radio galaxy NGC 6251 using observation data from the Fermi Large Area Telescope. The maximum likelihood analysis results show that a radio morphology template provides a better fit than a pointlike source description for the observational data at a confidence level of 8.1σ, and the contribution of lobes constitutes more than 50% of the total γ-ray flux. Furthermore, the γ-ray energy spectra show a significant disparity in shape between the core and lobe regions, with a curved log-parabola shape observed in the core region and a power-law form observed in the lobes. Neither the core region nor the northwest (NW) lobe displays significant flux variations in the long-term γ-ray light curves. The broadband spectral energy distributions of the core region and the NW lobe can be explained with a single-zone leptonic model. The γ-rays of the core region are due to the synchrotron-self-Compton process, while the γ-rays from the NW lobe are interpreted as inverse Compton emission of the cosmic microwave background.


Introduction
Radio galaxies (RGs) are a subclass of radio-loud (RL) active galactic nuclei (AGNs), characterized by the remarkable and complex large-scale jet morphology in the radio band.The largescale jet can extend beyond the host galaxy of RGs and form some substructures, such as knots, hotspots, and radio lobes, which have been resolved in the radio, optical, and X-ray bands (see Harris & Krawczynski 2006 for a review).So far, dozens of RGs have been detected by the Fermi Large Area Telescope (Fermi-LAT; Abdollahi et al. 2022;Ballet et al. 2023).Several RGs have even been detected at the TeV band and confirmed to be very high-energy emitters. 4Generally, the γ-rays of RGs are still considered to come from the core jet (Fukazawa et al. 2015;Xue et al. 2017).However, some works studying the X-ray emission of the large-scale jet substructures in RGs predicted the detectable γ-rays of some substructures (e.g., Zhang et al. 2009Zhang et al. , 2010Zhang et al. , 2018;;He et al. 2023).Especially, the detection of γ-rays from large-scale radio lobes of RGs Cen A (Abdo et al. 2010a;Sun et al. 2016) and Fornax A (Ackermann et al. 2016) confirms that the substructures of the jet are acceleration sites of high-energy particles.Hence, the origin of γ-ray emission for RGs is still debated.
NGC 6251 is a nearby (z = 0.0247, Wegner et al. 2003) giant RG with a radio angular scale extending to about 1°.2.Its radio morphology mainly consists of a radio core, a narrow straight jet with a roughly uniform cone angle, and two large-scale radio lobes (Waggett et al. 1977;Saunders et al. 1981).The jet starts within 1 pc of the nucleus, extends toward the northwest (NW) for approximately ¢ 4.5, and subsequently forms the NW lobe about 0°.3 away from the radio core (Saunders et al. 1981;Perley et al. 1984).The radio intensities of the jet and NW lobe are clearly higher than that of the southeast (SE) lobe, which is separated from the radio core by an angular distance of 0°.7 (Perley et al. 1984;Cantwell et al. 2020).The X-rays from the nucleus and its ~¢ 4.5 jet were initially observed by ROSAT (Birkinshaw & Worrall 1993).The diffuse X-ray emission originating from the NW lobe in NGC 6251 has been detected through the Suzaku X-ray imaging observations, which has been suggested to be generated by the inverse Compton scattering of cosmic microwave background (IC/CMB) process (Takeuchi et al. 2012).The X-ray emission from the straight jet region was also reported to potentially arise from the IC/CMB process (Sambruna et al. 2004), although an alternative explanation could be synchrotron radiation (Evans et al. 2005).Gamma-rays from NGC 6251 were likely first detected by the Compton Gamma-Ray Observatory in the MeV-GeV range.Using the X-ray images obtained from ROSAT and ASCA observations, Mukherjee et al. (2002) proposed that the Energetic Gamma-Ray Experiment Telescope source 3EG J1621+8203 is likely associated with NGC 6251.This association was further substantiated by Foschini et al. (2005) through an analysis of a wide field of view observation from INTEGRAL/IBIS.The GeV γ-ray emission of NGC 6251 has been reported in the first Fermi-LAT catalog (Abdo et al. 2010b), identified as a γ-ray emitting misaligned AGN (Abdo et al. 2010c), and subsequently included in the second/ third/fourth Fermi-LAT catalogs (Nolan et al. 2012;Acero et al. 2015;Abdollahi et al. 2022;Ballet et al. 2023).The observed γ-rays are conventionally believed to originate from the core-jet region near the black hole via the synchrotron-self-Compton (SSC) process (Chiaberge et al. 2003;Migliori et al. 2011;Xue et al. 2017).However, the 95% error ellipse of its associated Fermi-LAT source 2FGL J1629.4+8236does not encompass the radio core of NGC 6251 (Takeuchi et al. 2012).Additionally, the corresponding coordinates of this Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Fermi-LAT source in the latest fourth Fermi-LAT source catalog fall within the region of the jet in NGC 6251 at a distance of ~¢ 3 6 from the radio core (Foschini et al. 2022).These results make the origin of its γ-rays ambiguous.In this paper, we completely analyze the ∼15 yr Fermi-LAT observation data to investigate the origins of its γ-ray emission.The data analysis and results of Fermi-LAT are presented in Section 2. The spectral energy distributions (SEDs) of the NW lobe and core region are constructed and modeled in Section 3. A discussion and conclusions are given in Section 4. Throughout, H 0 = 71 km s −1 Mpc −1 , Ω m = 0.27, and Ω Λ = 0.73 are adopted in this paper.

Baseline Model of Data Analysis
NGC 6251 has been reported to be associated with a γ-ray source 4FGL 1630.6+8234 in the Fermi-LAT 12 yr Source Catalog (4FGL-DR3, Abdollahi et al. 2022). 5The Pass 8 data covering 2008 August 4 to 2023 March 1 (MJD 54682-60004) within the energy range of 0.1-300 GeV were downloaded from the Fermi Science Support Center. 6We select data within the 15°region of interest (ROI) centered on the radio position of NGC 6251 (R.A. = 248°.133, decl.= 82°.538).The publicly available software fermitools (v2.2.0, Fermi Science Support Development Team 2019) and the binned likelihood analysis method are used for our analysis.We use event class "SOURCE" (evclass=128) and event type "FRONT+BACK" (evtype=3) for the data analysis based on LAT data selection recommendations.To eliminate the contamination of γ-rays from the Earth's limb, the maximum zenith angle of 90°is set.A standard filter expression "(DATA_QUAL>0)&&(LAT_-CONFIG==1)" and the instrument response function of P8R3_SOURCE_V3 are used.All sources included in 4FGL-DR3 within ROI are added to the model.The spectral parameters of the sources encompassed in the circle centered on 4FGL J1630.6+8234 with a radius of 6°are left free, whereas the parameters of those sources lying beyond 6°are fixed to their 4FGL-DR3 values.The background models, including isotropic emission ("iso_P8R3_SOURCE_V3_V1. txt") and the diffuse Galactic interstellar emission ("gll_iem_v07.fits"), are considered and only their normalization parameters are kept free.
We use the maximum test statistic (TS) to evaluate the significance of the γ-ray signals of a source in the background, ) , where  src and  null are the likelihood values for the background with and without a source.If TS 25, it is considered that there is a new source (Abdo et al. 2010b).By generating a 5°× 5°residual TS map centered on the radio core of NGC 6251 for new source search, the maximum TS value of ∼10 is obtained, indicating that no new sources appear in the background.Generally, the power-law (PL) or log-parabola (LP) functions are used to fit the spectrum of sources, i.e., = , where N(E) is the photon distribution as a function of energy, Γ γ is the photon spectral index, E b is the scale parameter of photon energy, and β is the curvature parameter (Massaro et al. 2004).Following the methodology outlined in the second Fermi-LAT catalog (Nolan et al. 2012) ( ) to estimate the curvature significance of an energy spectrum.If TS curve 16 (corresponding to 4σ), it indicates that the spectrum exhibits significant curvature, and therefore, we use the LP function to describe it.

Pointlike Source Analysis
In 4FGL-DR4 (Abdollahi et al. 2022;Ballet et al. 2023), 4FGL J1630.6+8234 is identified as a pointlike source (PS) with an LP spectrum.Reanalyzing its ∼15 yr Fermi-LAT observation data, the TS map also shows it as a PS with TS ∼ 2619.34.Its time-integrated spectrum of the ∼15 yr Fermi-LAT observations within the energy band of 0.1-300 GeV is well explained by an LP spectral form with Γ γ = 2.36 ± 0.03 and β = 0.04 ± 0.02, as depicted in Figure 1 and shown in Table 1.The ∼15 yr average flux is F 0.1-300 GeV = (13.40± 0.51) × 10 −12 erg cm −2 s −1 .These results are roughly consistent with that of Γ γ = 2.27 ± 0.04, β = 0.10 ± 0.02, and F 0.1-100 GeV = (13.10± 0.56) × 10 −12 erg cm −2 s −1 in 4FGL-DR4.The long-term light curve of 4FGL J1630.6+8234 is generated using the adaptive-binning method (Lott et al. 2012) to ensure TS 9 for each time bin, where the minimum bin size is 30 days.The ∼15 yr Fermi-LAT light curve of 4FGL J1630.6+8234 in the 0.1-300 GeV band demonstrates a steady state without any significant flux variation, as displayed in Figure 1.If TS < 4, an upper limit is presented for that energy bin.The light curve (in the right panel) is obtained with an adaptive-binning method based on a criterion of TS 9 for each time bin, where the minimum time bin step is 30 days.The horizontal red dashed line is the ∼15 yr average flux, i.e., F 0.1−300 GeV = (13.40± 0.51) × 10 −12 erg cm −2 s −1 .
Using the tool of gtfindsrc, we reestimate the best-fit position of 4FGL J1630.6+8234 and obtain (R.A. = 247°.752,decl.= 82°.558)with a 95% confidence error circle radius of 0°.015.This position is located in the NW region relative to the radio core of NGC 6251, as presented in Figure 2. Together with the steady γ-ray emission of 4FGL J1630.6+8234 in the past ∼15 yr indicates that the γ-ray emission of 4FGL J1630.6 +8234, at least a portion of γ-rays, is likely dominated by the radiation of the large-scale extended regions in NGC 6251.

Radio Morphology Template
In order to examine our speculation, we first generate the γray counts map of 4FGL J1630.6+8234 in the 0.1-300 GeV band, as depicted in Figure 2. The data reduction procedure is described in Appendix A.1.The radio contours of NGC 6251 at 609 MHz observed by the Westerbork Synthesis Radio Telescope (WSRT; Mack et al. 1997), the 95% error ellipses for the counterpart of NGC 6251 from 1FGL to 4FGL, and the 95% error circle of the best-fit position of this work are also presented in Figure 2. The confidence intervals of the best-fit position decrease with increasing LAT observation time, but the radio core of NGC 6251 is never encompassed.The distribution of γ-ray photons on the NW lobe of NGC 6251 is significantly prominent, while the brightness of the SE lobe appears relatively subdued, rendering it indistinguishable in the map.As shown in Figure 5 in Appendix B, the spatial  distributions of γ-ray photons exhibit significant variations across the low-energy to high-energy bands, shifting from a diffusion distribution dominated by the large-scale extended regions to a PS dominated by the radio core.The γ-ray photons above 10 GeV are primarily concentrated in close proximity to the radio core region.
Assuming that γ-rays are produced through the IC process by the same group of electrons responsible for generating radio radiation, the distribution of γ-rays can be considered a form of radio radiation (Ackermann et al. 2016).Therefore, using a Radio Morphology Template (RMT) model, we reanalyze the ∼15 yr Fermi-LAT observation data of 4FGL J1630.6+8234.We employ RMT to partition the observation area into three zones, namely, the core region and two giant lobes.The regions of two giant lobes are selected with the 609 MHz radio image from the NASA/IPAC Extragalactic Database (2019).Of particular note, the overlap region between the radio green contours and the red circle (Figure 5 in Appendix B) is subtracted and not included for the NW lobe template in the RMT model.The red circle is centered on the radio core with a radius of ¢ 4.5, which encompasses the radio core, a bright straight jet extending to the NW from the radio core, and the faint counterjet.The two lobes are taken as extended radiation sources while the core region is set as a PS model centered around the radio core of NGC 6251.We use the RMT model instead of the PS 4FGL J1630.6+8234 in the XML file and reperform the maximum likelihood analysis.
Through the RMT model analysis, we obtain the spectra and long-term light curves of lobes and core region, as shown in Figure 3.The spectrum of the core region still requires an LP function to fit with Γ γ = 1.80 ± 0.13 and β = 0.28 ± 0.06, which is more curved and harder than the total integrated spectrum of 4FGL J1630.6+8234(Figure 1).The ∼15 yr average flux of the core region in the 0.1-300 GeV band is F 0.1−300 GeV = (4.86 ± 0.61) × 10 −12 erg cm −2 s −1 with TS ∼508.55.The spectra of two lobes can be well explained with a PL form.We obtain Γ γ = 2.54 ± 0.06 and an average flux of F 0.1−300 GeV = (7.67 ± 0.76) × 10 −12 erg cm −2 s −1 with TS ∼600.31 for the NW lobe.The SE lobe displays a very steep spectrum in the Fermi-LAT energy band with Γ γ = 7.01 ± 0.19.The average flux F 0.1−300 GeV = (0.96 ± 0.23) × 10 −12 erg cm −2 s −1 and the TS value 30.96 are far lower than that of the core region and NW lobe.The results are shown in Table 1.Almost no emission above 1 GeV can be inspected in the SE lobe, which is consistent with the γ-ray counts maps as shown in Figure 5 in Appendix B. The long-term light curves of the core region and NW lobe under the case of the RMT model are simultaneously produced by fixing the parameters of the SE lobe and other sources.The adaptive-binning method is also used and a minimum time bin of 60 days is taken.Neither the NW lobe nor the core region shows significant flux variation of γ-rays, as illustrated in Figure 3 (see Appendix A.2).

PS Model versus RMT Model
Clearly, the TS value and flux of the giant NW lobe are higher than those of the core region.The γ-rays from lobes account for more than 50% of the total γ-ray flux of the source.We compare the likelihood values of the PS model with the RMT model and assess the significance by ), where the RMT model has four additional free parameters compared to the PS model.We They are all derived with the ∼15 yr Fermi-LAT observation data in the 0.1-300 GeV band.The solid red lines and gray dashed lines (in two left panels) depict the fitted spectra along with their respective 1σ uncertainties.If TS<4, an upper limit is presented for that energy bin.The light curves are derived with an adaptive-binning method based on a criterion of TS 9 for each time bin, where the minimum time bin step is 60 days.The data points of each time bin in the two light curves are simultaneously obtained.The horizontal red dashed lines represent the ∼15 yr average flux for each region, i.e., F 0.1−300 GeV = (4.86 ± 0.61) × 10 −12 erg cm −2 s −1 for the core region and F 0.1−300 GeV = (7.67 ± 0.76) × 10 −12 erg cm −2 s −1 for NW lobe.
obtain TS ext = 78.28,indicating that the RMT model is statistically favored over the PS model at a confidence level of 8.1σ.
As described above, the absence of significant variability in the overall long-term light curve of 4FGL J1630.6+8234(Figure 1) may be attributed to the dominance of lobe emission; however, no discernible flux variation is observed in the light curve of the core region either (Figure 3).In order to further study this issue, we derive the long-term light curve of NGC 6251 in the energy band of 10-300 GeV since the γ-rays above 10 GeV potentially are dominated by the emission of the core region.Nevertheless, only three detection points can even be obtained using the time bin of 720 days to analyze the observational data, and they do not show the obvious flux variation, as illustrated in Figure 6 in Appendix B.

Constructing and Modeling the SED
The distinct spectral patterns observed in the core region and NW lobe also suggest that γ-rays from 4FGL J1630.6+8234have separate origins.We construct the broadband SEDs for the NW lobe and core region, respectively, spanning from the radio to γ-ray bands, as depicted in Figure 4.For the core region, the radio and optical-UV data are collected from Migliori et al. (2011), X-rays are the combined spectrum using the latest 11 X-ray Telescope (XRT) observations (see Appendix B), and γ-rays are obtained by the analysis in this work.For the NW lobe, the radio and X-ray data are collected from Takeuchi et al. (2012) and Persic & Rephaeli (2019), while the γ-ray data are obtained by the analysis in this work.
We adopt a single-zone leptonic radiation model to fit the broadband SEDs of the NW lobe and core region.The model includes the synchrotron and SSC (and IC/CMB) processes of the relativistic electrons in radiation regions.The emission region is assumed to be a sphere with a radius R, magnetic field strength B, the bulk Lorenz factor Γ, and the Doppler boosting factor δ. The electron distribution is assumed as a broken PL function, characterized by an electron density parameter N 0 , a break energy γ b , and indices p 1 and p 2 in the range of g min [ , g max ].The synchrotron-self absorption, Klein-Nishina effect, and the extragalactic background light (EBL) absorption (Franceschini et al. 2008) are also taken into account during the SED modeling.

Core Region
We only consider the synchrotron and SSC processes to reproduce the observed SED of the core region, the same model as the previous works for NGC 6251 (Chiaberge et al. 2003;Migliori et al. 2011;Xue et al. 2017).The size of the radiation region is determined as R = δcΔt/(1 + z), where Δt is set to 10 days and c represents the speed of light.The values of p 1 and p 2 are obtained based on the spectral indices at the radio/X-ray and γ-ray bands, respectively.g min cannot be constrained and is set as 200, while g max is roughly constrained by the last data point in the γ-ray band.We adjust the parameters B, δ, γ b , and N 0 to represent the observed SED of the core region.It should be noted that these parameter values may not be unique, in particular, B and δ are degenerate (Zhang et al. 2012).The fitting parameters are given in Table 2.
The model can well reproduce the observed SED of the core region, as illustrated in Figure 4. We obtain δ = 3.4, which is lower than the typical values observed in blazars, coinciding with the uniform model of RL AGNs (Urry & Padovani 1995).The same model has been previously employed to explain the archival SEDs of the core region for NGC 6251, where B = 0.03 G and δ = 3.2 were reported (Chiaberge et al. 2003), as well as B = 0.037 G and δ = 2.4 (Migliori et al. 2011).Our The magenta, cyan, and green dashed lines represent the synchrotron, SSC, and IC/CMB components, respectively.The black solid lines show the sum of each emission component.For the core region, the data from radio to optical-UV (gray circles) are taken from Migliori et al. (2011), where the gray solid circles indicate the data corrected the dust extinction.The olive circles are the Swift-XRT data obtained in this work.The red symbols represent the Fermi-LAT data and are the same as that in Figure 3.For the NW lobe, the radio (gray opened circles) and X-ray (brown opened circles) data are taken from Takeuchi et al. (2012) and Persic & Rephaeli (2019).The red symbols represent the analysis in this work and are taken from Figure 3. 6.4 × 10 3 3 × 10 5 g max 1.9 × 10 5 3.5 × 10 6 N 0 (cm −3 ) 3.8 × 10 5 7 × 10 −7 p 1 2.64 2.4 p 2 3.52 4 findings of B = 0.04 G and δ = 3.4 are roughly consistent with theirs.

NW Lobe
We employ the synchrotron+SSC+IC/CMB processes to fit the broadband SED of the NW lobe.Considering no relativistic motion for the giant lobe, δ = Γ = 1 is adopted.In the comoving frame, the CMB energy density is given by & Schlickeiser 1994;Georganopoulos et al. 2006), where U CMB = 4.2 × 10 −13 erg cm −3 .The radius of the radiation region is taken as R = 185 kpc (Takeuchi et al. 2012).The values of p 1 and p 2 are determined from the spectral indices at radio and γ-ray bands, respectively.g min cannot be constrained and is set as 100 while g max is roughly constrained by the last data point in the γray band.We adjust the values of parameters B, γ b , and N 0 to reproduce the observed data.The fitting parameters are provided in Table 2.
As depicted in Figure 4, the emission from the X-ray to γ-ray bands is dominantly produced by the IC/CMB process.In comparison to the IC/CMB component, the contribution from SSC is negligible.The previous study utilized the IC/(CMB +EBL) model, employing a double broken PL electron distribution to represent the X-ray data of the NW lobe in NGC 6251 (Takeuchi et al. 2012), and they yielded B = 0.37 μG and g = 10 , and the slope index of q = 2.5.Our results are consistent with theirs.

Discussion and Conclusions
The RG NGC 6251 is reported to be associated with the γray source 4FGL J1630.6+8234, and 4FGL J1630.6+8234 is identified as a PS in 4FGL-DR4 (Abdollahi et al. 2022;Ballet et al. 2023).We comprehensively analyze the ∼15 yr Fermi-LAT observation data of 4FGL J1630.6+8234 and observe that the giant lobes dominate the detected γ-rays of NGC 6251.By comparing the maximum likelihood values between the RMT model analysis and a PS model analysis, we find that the RMT model analysis is preferable to describe the γ-ray emission of NGC 6251 at a confidence level of 8.1σ.
The spectrum of the core region is better described by an LP function compared to a PL function, while the spectra of the two lobes can be well explained with a PL form.The distinct spectral shapes between the core region and lobes also demonstrate the different origins of γ-rays.The γ-ray flux of the NW lobe is even higher than that of the core region, accounting for more than 50% of the total γ-ray flux.However, the emission flux from the SE lobe is very low, only being ∼7% of the total flux.The TS value of the SE lobe is only 30.96, while it is TS ∼600.31 for the NW lobe and TS ∼508.55 for the core region.
Neither the core nor the NW lobe displays significant flux variation in their long-term γ-ray light curves.However, X-ray observations of NGC 6251 using Swift-XRT indeed show flux variations at a confidence level of 4.9σ (Figure 6 in Appendix B), which are likely attributed to the core-jet emission.It is worth noting that the core region in the RMT model encompasses the straight jet; the low-level γ-ray flux variations originating from the radio core might be concealed by the radiation of the straight jet.This further supports the conclusion that the dominant contribution to γ-ray fluxes observed from 4FGL J1630.6+8234stems from the extended jet substructures of NGC 6251, surpassing that of the radio core.
On the basis of the derived average spectra in the X-ray and γ-ray bands, we construct broadband SEDs of the NW lobe and core region and fit them with a one-zone leptonic model.The γrays of the NW lobe together with its X-ray emission are attributed to the IC/CMB by the relativistic electrons, while the γ-rays of the core region are produced by the SSC process.The derived parameter values of radiation regions are consistent with previous works that studied NGC 6251.
Following the examples of Cen A (Abdo et al. 2010a) and Fornax A (Ackermann et al. 2016), NGC 6251 is the third RG with measured extended γ-ray emission.Our findings would further improve the cognition of the diversity γ-ray emitters in the γ-ray band.The γ-rays from the large-scale extended regions of the three RGs can be measured only due to their close proximity and the large angular size.The low spatial resolution of γ-ray detectors makes it difficult to judge the location of the γ-ray emission for most sources.On this point, the γ-ray emission from the large-scale jets of RGs may be universal, as predicted by some theoretical works (Zhang et al. 2009(Zhang et al. , 2010(Zhang et al. , 2018)).On the other hand, most detected extragalactic sources in the GeV-TeV γ-ray band are blazars due to their aligned jets with the strong Doppler amplification of their emission.The lack of the relativistic effect in the largescale jets implies that the detected γ-rays from these regions are intrinsically strong.Therefore, the large-scale jets of AGNs may provide more energy input to the intergalactic medium in the γ-ray band (H.E. S. S. Collaboration et al. 2020).

A.2. Estimation of the γ-Ray Variability
To quantify the variability of γ-ray light curves, we calculate χ 2 and the associated probability p(χ 2 ) = 1 − p(>χ 2 ; Chen et al. 2022 and references therein) of the long-term γ-ray light curves, å å where N is the number of the data points, 〈F〉 is the weighted mean flux, and F i and σ i are the flux and its error for the ith data point.However, none of the γ-ray light curves presented in this paper exhibits significant variability at a confidence level exceeding 2σ.

Appendix B Swift-XRT Observations and Data Analysis
The XRT on board the Neil Gehrels Swift Observatory (Swift, Gehrels & Swift 2004;Burrows et al. 2005) has performed a total of 18 observations for NGC 6251 in the photon counting readout mode, spanning from 2007 April to 2023 October.One observation was unable to capture the radio position of NGC 6251, due to it being outside the view field of Swift-XRT.For two other observations, poor quality data were obtained.Therefore, our analysis is based on a total of 15 observation data, as listed in Table 3.The data are processed using the XRTDAS software package (v.3.7.0),where the software package was developed by the ASI Space Science Data Center and released by the NASA High Energy Astrophysics Archive Research Center in the HEASoft package (v6.30.1).The calibration files from XRT CALDB (v20220803) are used within xrtpipeline to calibrate and clean the events.The individual XRT event files are merged together using the xselect package, and then the average spectrum is created from the combined event file.Events for the spectral analysis are extracted from a circle centered on the radio core of NGC 6251 with a radius of 20″.The background is taken from an annulus with an inner and outer radii of 40″ and 80″.The ancillary response files, which are applied to correct the pointspread function losses and CCD defects, are generated with the xrtmkarf task using the cumulative exposure map.The spectrum is grouped to ensure at least 20 counts per bin and the χ 2 minimization technique is adopted for spectral analysis.The spectrum is fitted by a single PL with two absorption components, one is absorption at z = 0 with the neutral hydrogen column density fixed at Galactic value of We generate the XRT spectra at other epochs following the procedures mentioned above.The derived photon index (Γ X ) and corrected flux (F 0.5-10 keV ) for the 15 observational epochs are given in Table 3.The X-ray light curve obtained with the XRT observations is presented in Figure 6.Using Equations (A1) and (A2) above, we obtain 〈F〉 0.5-10 keV = 2.50 × 10 −12 erg cm −2 s −1 , χ 2 = 57.4(d.o.f = N − 1 = 14) with p = 3.4 × 10 −7 , indicating that there is variability of X-rays at a confidence level of 4.9σ.
We also produce a combined average spectrum by incorporating the latest 11 observational data spanning from 2022 September to 2023 October, which are all corrected values, as displayed in Figure 4.It is worth noting that the first three Swift-XRT observational data were previously reported by Migliori et al. (2011), and our findings are consistent with them within the errors.

Figure 1 .
Figure1.Spectrum and light curve of 4FGL J1630.6+8234 when considering it as a PS.They are all derived with the ∼15 yr Fermi-LAT observation data in the 0.1-300 GeV band.The solid red line and gray dash lines (in left panel) represent the fitting result of the spectrum and the corresponding 1σ uncertainty, respectively.If TS < 4, an upper limit is presented for that energy bin.The light curve (in the right panel) is obtained with an adaptive-binning method based on a criterion of TS 9 for each time bin, where the minimum time bin step is 30 days.The horizontal red dashed line is the ∼15 yr average flux, i.e., F 0.1−300 GeV = (13.40± 0.51) × 10 −12 erg cm −2 s −1 .

Figure 2 .
Figure 2. 1°. 5 × 1°. 5 γ-ray counts map of 4FGL J1630.6+8234 in the 0.1-300 GeV band.The color bar represents the counts in each pixel with a pixel size of 0°.01.The map has been smoothed with a Gaussian kernel of 0°.25.The green contours denote the large-scale radio structure of NGC 6251 obtained with the WSRT observations (Mack et al. 1997).The black cross represents the position of the radio core in NGC 6251.The red circle indicates the 95% error circle of the reestimated best-fit position for 4FGL J1630.6+8234obtained with the ∼15 yr Fermi-LAT observation data in this paper.The 95% error ellipses of 4FGL J1630.6+8234from 1FGL to 4FGL (marked respectively with gray, brown, blue, and black ellipses) are also presented for comparison.

Figure 3 .
Figure3.Spectra and light curves of the core region (the top panels) and NW lobe (the bottom panels) for NGC 6251.They are all derived with the ∼15 yr Fermi-LAT observation data in the 0.1-300 GeV band.The solid red lines and gray dashed lines (in two left panels) depict the fitted spectra along with their respective 1σ uncertainties.If TS<4, an upper limit is presented for that energy bin.The light curves are derived with an adaptive-binning method based on a criterion of TS 9 for each time bin, where the minimum time bin step is 60 days.The data points of each time bin in the two light curves are simultaneously obtained.The horizontal red dashed lines represent the ∼15 yr average flux for each region, i.e., F 0.1−300 GeV = (4.86 ± 0.61) × 10 −12 erg cm −2 s −1 for the core region and F 0.1−300 GeV = (7.67 ± 0.76) × 10 −12 erg cm −2 s −1 for NW lobe.

Figure 4 .
Figure4.Broadband SEDs of the core region (left panel) and NW lobe (right panel) along with the fitting results.The magenta, cyan, and green dashed lines represent the synchrotron, SSC, and IC/CMB components, respectively.The black solid lines show the sum of each emission component.For the core region, the data from radio to optical-UV (gray circles) are taken fromMigliori et al. (2011), where the gray solid circles indicate the data corrected the dust extinction.The olive circles are the Swift-XRT data obtained in this work.The red symbols represent the Fermi-LAT data and are the same as that in Figure3.For the NW lobe, the radio (gray opened circles) and X-ray (brown opened circles) data are taken fromTakeuchi et al. (2012) andPersic & Rephaeli (2019).The red symbols represent the analysis in this work and are taken from Figure3.

max 6 (
corresponding to the value of γ b in our work).Using a similar model with a PL electron distribution to fit the SED, Persic & Rephaeli (2019) obtained B = 0.40 μG, g = 1.1 10 max 6

Figure 5 .
Figure5.The 1°. 5 × 1°. 5 γ-ray counts map of 4FGL J1630.6+8234 in different energy bands.The top left, top right, bottom left, and bottom right panels are in the energy bands of 0.1-3 GeV, 1-3 GeV, 3-10 GeV, and 10-300 GeV, respectively.The symbols remain unchanged from those in Figure2, with the exception of the red circle.Here, the red circle represents a region centered on the position of the radio core with a radius of ¢ 4. 5. Note that the red circle is subtracted and not included for the NW lobe template in the RMT model.

Figure 6 .
Figure 6.Top panel: the ∼15 yr γ-ray light curve in the 10-300 GeV band of 4FGL J1630.6+8234 when considering it as a PS, each time bin is 720 days and the triangles indicate TS <9 for this time bin.Bottom panel: the X-ray light curve of NGC 6251, obtained by the Swift-XRT observation data.The horizontal red dashed line represents the average flux of all the data points, 〈F〉 0.5-10 keV = 2.50 × 10 −12 erg cm −2 s −1 .

Table 1
Test Results for Different Spatial Templates aThe LP spectral form.bThePL spectral form.

Table 2
Fitting Parameters of SED Fitting