Examining a hadronic $γ$-ray scenario for the radiative shell & molecular clouds of the old GeV supernova remnant G298.6$-$0.0

TL;DR

This study refines Fermi-LAT gamma-ray measurements for the old SNR G298.6$-$0.0, applying higher-resolution PSF-domain data to better separate low-energy flux and reduce contamination from the Galactic diffuse background. By adopting a hadronic framework with a proton population featuring a minimum energy, spectral index, and an exponential cutoff, the authors decompose the emission into Src-NE (shell-dominated) and Src-NW (farther MCs illuminated by escaped CRs), with Src-S likely involving additional sources. BKPL fits reveal spectral breaks at $E_{\mathrm{br}}=1.50^{+0.60}_{-0.50}$ GeV for Src-NE and $E_{\mathrm{br}}=0.68^{+0.32}_{-0.11}$ GeV for Src-NW, and the derived CR populations yield $E_{\mathrm{CR,min}}$, $\Gamma_{\mathrm{CR}}$, and $E_{\mathrm{CR,max}}$ consistent with shell and MC scenarios; Src-NE shows a relatively soft CR spectrum and a sub-GeV to GeV cutoff, while Src-NW requires a higher $E_{\mathrm{CR,min}}$ and no detectable cutoff. The energy budget, diffusion considerations, and comparisons with W28 and W44 constrain the remnant’s age to $\sim$10–30 kyr, reinforcing the view that old SNRs remain important laboratories for studying CR escape and SNR–MC interactions. Overall, the work demonstrates a robust, hadronic-based method to disentangle shell and MC gamma-ray components in complex SNR environments and tightens the case for G298.6$-$0.0 as an old ($>10$ kyr) remnant with distinct shell and MC contributions.

Abstract

Based on the 13.7~yr Fermi-LAT data, Yeung et al. (2023) claimed detection of two $γ$-ray sources (Src-NE and Src-NW) associated with the supernova remnant (SNR) G298.6$-$0.0, and interpreted it as an old GeV SNR interacting with molecular clouds (MCs). In this follow-up study, we refine the flux measurements below 2~GeV with Fermi-LAT event types of better angular reconstruction. Then, we report our cosmic-ray phenomenology in a hadronic scenario, considering both the shell and MC regions of SNR G298.6$-$0.0. We confirm that both the $γ$-ray spectra of Src-NE and Src-NW exhibit spectral breaks at $1.50_{-0.50}^{+0.60}$~GeV and $0.68_{-0.11}^{+0.32}$~GeV, respectively. Src-NW has a harder broadband photon index than Src-NE, suggesting an appreciable difference between the physical separations of their respective emission sites from SNR G298.6$-$0.0. The cosmic-ray spectrum responsible for Src-NE starts with a minimum energy $E_\mathrm{CR,min}=1.38_{-0.16}^{+0.47}$~GeV, and has a proton index $Γ_\mathrm{CR}=2.57_{-0.21}^{+0.18}$ below the exponential cutoff energy $E_\mathrm{CR,max}=240_{-150}^{+240}$~GeV. Accordingly, we argue that Src-NE is dominated by the SNR shell, while only a minor portion of lower-energy emission is contributed by the MCs interacting with the SNR. The cosmic-ray population for Src-NW starts at a higher energy such that the $E_\mathrm{CR,min}$ ratio of Src-NW to Src-NE is $\gtrsim$2. The high $E_\mathrm{CR,min}$, as well as the high cosmic-ray energy density required ($\sim$26~eV~cm$^{-3}$), supports the interpretation that Src-NW is predominantly the $γ$-ray emission from the farther MCs being bombarded by protons that had earlier escaped from SNR G298.6$-$0.0. By comparing the high-energy features of G298.6$-$0.0 with those of analogical SNRs, especially SNR W28 and SNR W44, we further constrain the age of SNR G298.6$-$0.0 to be 10--30~kyr.

Figures (5)

• Figure 1: The spatial configuration of the SNRs, MC clumps and $\gamma$-ray components in our targeted region. The MOST 843 MHz radio continuum contours of SNR G298.6$-$0.0 and SNR G298.5$-$0.3 Whiteoak1996 are plotted in blue. Three clumps of dense MCs at approximately the same distance from us as G298.6$-$0.0 ($\sim$10.1 kpc), which are traced by $^{12}$CO($J$=1--0) (115 GHz) line emission (Figure 5 of Yeung2023), are marked as red crosses. The approximate size of the clump MC-W is indicated by the red circle. The Fermi-LAT $\gamma$-ray centroids of Src-NE, Src-NW and Src-S are marked as black diamonds encircled by their respective error circles at the 95% confidence level (taken from Figure 1 of Yeung2023).
• Figure 2: The $\gamma$-ray spectral energy distribution of Src-NE (top), and the residual (data$-$model) divided by the corresponding combined uncertainty when fitting the cosmic-ray phenomenological model to the plotted data (bottom). PSF1+PSF2+PSF3 data below 2 GeV (red filled circles) and FRONT+BACK data above 2 GeV (blue open squares) are adopted for Chi-Square fittings of models (black lines). The combined uncertainties, each of which is the statistical and systematic uncertainties added in quadrature, are plotted in a black-dotted style. Upper limits at the 95% confidence level are marked as black crosses with a down arrow. The red dotted line is a model for crosschecking, which is reconstructed with a maximum-likelihood fit to the PSF1+PSF2+PSF3 data in 0.3--100 GeV. The grey dotted lines in the bottom panel mark the values of 0 and $\pm$2. A partial duplicate of this figure is presented in Figure \ref{['apx']}, supplemented with some crosschecking information.
• Figure 3: The $\gamma$-ray spectral energy distribution of Src-NW (top), and the residual (data$-$model) divided by the corresponding combined uncertainty when fitting the cosmic-ray phenomenological model to the plotted data (bottom). The colours, symbols and linestyles have the same meanings as in Figure \ref{['SED_Src-NE']}.
• Figure 4: The $\gamma$-ray spectral energy distribution of Src-S (top), and the residual (data$-$model) divided by the corresponding combined uncertainty when fitting the cosmic-ray phenomenological model to the plotted data (bottom). The colours, symbols and linestyles have the same meanings as in Figure \ref{['SED_Src-NE']}.
• Figure 5: The $>$2 GeV binned spectra of Src-NE (a partial duplicate of Figure \ref{['SED_Src-NE']}). Red filled circles are reconstructed with PSF1+PSF2+PSF3 data, and blue open squares are reconstructed with FRONT+BACK data Yeung2023. The error bars plotted here are purely statistical. Dark-yellow diamonds represent the altered fluxes in response to shifting the normalisation of the Galactic diffuse model by $\pm$5%, while the grey diamonds represent the altered fluxes in response to simultaneously shifting the normalisations of Src-NW and Src-S by $\pm$10%.