EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 705 (January 2026) A&A, 705 (2026) A107 Full HTML
Open Access
Issue
A&A
Volume 705, January 2026
Article Number A107
Number of page(s) 7
Section Astrophysical processes
DOI https://doi.org/10.1051/0004-6361/202555650
Published online 09 January 2026

© The Authors 2026

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1. Introduction

The magnetosphere and surroundings of rapidly rotating neutron stars, i.e., pulsars, are ideal astrophysical environments for accelerating particles up to GeV–PeV energies and are observed as bright sources of nonthermal photon emission from radio to TeV energies (Philippov & Kramer 2022). Specifically, electron and positron pairs (e±) are created in electromagnetic cascades in the pulsar’s magnetosphere and can be further accelerated, for example, in the termination shock of the relativistic pulsar wind. Multiwavelength signatures of these processes are accessible through the observation of pulsars’ spectra, their light curves, and the extended pulsar nebula (PWN) emission Gaensler & Slane (2006). Furthermore, after being accelerated, the e± are released into the interstellar medium and likely contribute to the cosmic ray density of our Galaxy, providing a primary antimatter component of positrons to the local cosmic-ray fluxes (see e.g. Manconi et al. 2020; Evoli et al. 2021; Orusa et al. 2025). An important confirmation of this idea is the observation of a few-degree extended gamma-ray emission around one of the closest and most powerful pulsars, the Geminga pulsar (López-Coto et al. 2022; Liu 2022; Fang 2022; Amato & Recchia 2024). When scaled to the source distance (about 250 pc), this demonstrates that e± populate a halo of a few tens of parsecs around the central pulsar. These particles produce gamma rays through inverse Compton emission on the interstellar radiation fields and have escaped the PWN. This phenomenon, called a “TeV halo” or simply a “pulsar halo”, is supported by observations from Milagro, HAWC, HESS, LHAASO, and Fermi-LAT (Abdo et al. 2009; Abeysekara et al. 2017; Aharonian et al. 2023, 2021; Di Mauro et al. 2019) for Geminga and a few other sources at TeV energies (Aharonian et al. 2021). The e± populating such halos are expected to produce a complementary signature at X-ray energies, through synchrotron emission, via interaction with the ambient magnetic field. However, searches for the X-ray synchrotron counterpart of the GeV–TeV emitting e± have so far yielded null results (Manconi et al. 2024; Khokhriakova et al. 2024; Adams et al. 2025), resulting in strong constraints on the ambient magnetic field strength. Current X-ray observations do not cover the full angular extension of pulsar halos, specifically the observed Geminga emission at TeV energies. The typical field of view (FOV) of current X-ray observatories is limited to a few tens of arcseconds around Geminga and can be extended to a few degrees only in the hard X-ray band with NuSTAR by accounting for stray-light photons (Manconi et al. 2024). For these reasons, a wide-field search for the X-ray counterpart of the Geminga halo emission offers a complementary view and could potentially be more sensitive to the large-scale, diffuse X-ray counterpart of TeV pulsar halos.

Recently, X-ray emission at 0.2–2 keV, extending up to 0.2 degrees, has been reported from the putative synchrotron halo of the Monogem pulsar using SRG/eROSITA performance verification data obtained in pointing mode (Niu et al. 2025); see, however, Khokhriakova et al. (2025). This detection would be consistent with previous null detections of the same object with XMM-Newton and SRG/eROSITA surveys (Liu et al. 2019; Khokhriakova et al. 2024). The properties of the observed emission provide information on the strength and spatial structure of the underlying magnetic field, hinting at a nonuniform magnetic field structure. These results demonstrate that wide FOV observations are a promising avenue for elucidating particle acceleration and transport in pulsar halos.

In this paper, we search for the putative X-ray halo of the Geminga pulsar using SRG/ART-XC observations, covering an area of the sky comparable in size to the expected Geminga emission. These data provide a unique wide-field dataset compared with previous narrow FOV studies. Dedicated SRG/ART-XC observations of Geminga complement the SRG/eROSITA study by Khokhriakova et al. (2024), providing harder energy coverage that is less biased by Galactic absorption and by contributions from soft diffuse emission (mainly nearby Monogem Ring supernova remnant, Plucinsky et al. 1996). Despite the non-detection of the Geminga pulsar halo in the current one-day ART-XC observation, we demonstrate that an additional exposure can provide a potential detection or stronger constraints on Geminga’s ambient magnetic field.

Our paper is organized as follows. Sect. 2 describes a benchmark model for the spectral and angular properties of the Geminga halo, from GeV to TeV energies, and the synchrotron emission from the pulsar halo in the X-ray band. In Sect. 3, we provide details of the SRG/ART-XC scanning observations, data reduction, and spatial fitting procedure. Section 4 presents the resulting upper limit on the Geminga halo detection. Section 5 describes simulations estimating the ability of SRG/ART-XC to improve the upper limit obtained in the current observation. Section 6 summarizes the constraints on the degree-wide Geminga pulsar diffuse emission with SRG/ART-XC and provides suggestions for future observations.

2. Pulsar halo model

To model the Geminga pulsar halo, we followed the framework presented in Manconi et al. (2024), to which the reader is referred for further details. Specifically, the benchmark adopted here was tuned to describe the spectral and angular properties of gamma-ray observations of the Geminga halo from GeV to TeV energies. Its main components are described below.

The e± are produced and accelerated up to multi-TeV energies continuously at a rate that follows the evolution of the pulsar’s spin-down luminosity, E ˙ ( t ) = E ˙ 0 ( 1 + t τ 0 ) − n + 1 n − 1 $ \dot{E}(t) = \dot{E}_0 ( 1+ \frac{t}{\tau_0})^{-\frac{n+1}{n-1}} $. We assumed magnetic dipole braking with index n = 3, fixing the typical decay time τ0 = 12 kyr and the current spin-down power to 3.2 × 1034 erg s−1. The injection spectrum of the accelerated e± was modeled using a power law with slope γ = 1.85 and an exponential cutoff Ec = 200 TeV of the form Q(E) = Q0E−γexp(−E/Ec), where normalization Q0 is related to the total rotational energy W0 through a conversion efficiency η = 0.12 as

∫ 0 T d t ∫ 0.1 GeV ∞ d E E Q ( E , t ) = η W 0 , $$ \begin{aligned} \int _{0}^T dt \int _{0.1 \mathrm {GeV}}^{\infty } dE \, E Q(E,t) =\eta W_0\, , \end{aligned} $$(1)

in which T = 342 kyr is the characteristic age of the pulsar (see Equations (9)–(12) from Di Mauro et al. 2019).

The choice of the injection spectral index γ is consistent with the photon indexes of the X-ray synchrotron radiation measured by Chandra in the axial tail A region of Geminga (see Posselt et al. 2017, for details), which may trace the escaping accelerated pairs. Indeed, e± pairs accelerated in the Geminga PWN could escape into the halo from this region, where photon indices of 1.4−1.6 measured by Chandra are produced by accelerated pairs with a power-law momenta distribution of index γ = 1.8 − 2.2.

After acceleration, e± begin to propagate in the pulsar surroundings and are subject to diffusion and energy losses, mainly by synchrotron and inverse Compton radiation in the ambient magnetic and photon fields, respectively. Inverse Compton energy losses were computed using the interstellar radiation field from Vernetto & Lipari (2016). This process produces photons at GeV–TeV energies. The same population of energetic e± pairs is expected to emit synchrotron X-rays in the keV band. This lower-energy counterpart of extended gamma-ray halos remains undetected to date, but its observation is crucial, for example, to constrain the properties of the ambient magnetic field in which e± propagate, as well as high-energy particle propagation models. For simplicity, the synchrotron emission was computed considering a uniform, random ambient magnetic field of strength B extending at least to the scale of the observed gamma rays. This approach provided a means to explore values compatible with, or higher than, the upper limits derived in Manconi et al. (2024), since the full angular extent of a possible X-ray counterpart remains unexplored.

The approximation of a uniform random field may not apply if the large-scale magnetic field around Geminga has a complicated spatial structure, for example, in the case of anisotropic diffusion (Liu 2022). This could affect the properties of the predicted synchrotron emission, such as the angular distribution of the X-rays around the source, as seed e± would diffuse faster along the mean direction of the magnetic field, as quantified in Wu et al. (2024). However, these modifications are strongly model dependent and are expected to give rise to differences in the normalization of the signal that are likely degenerate with other model parameters. In the absence of a robust synchrotron halo detection with X-rays, we considered the uniform magnetic field case for simplicity, leaving more complicated structures to further investigations.

The morphological and spectral properties of the nonthermal emission of energetic e± strongly depend on how particles are injected and transported in the vicinity of the pulsar. Currently, three main phenomenological models have been proposed to explain the properties of observed gamma rays, namely suppressed diffusion, ballistic propagation, and anisotropic diffusion (see Liu 2022 for a review). We considered the first scenario, in which particle diffusion is described by a diffusion coefficient of the form D(E) = D0(E/1 GeV)δ, where δ = 0.33 and D0 = 1026 cm2 s−1, suppressed by a factor of about 500 relative to the mean value inferred for the rest of the Galaxy. Apart from being the benchmark scenario used in much of the literature, suppressed diffusion can successfully explain the observed properties of the Geminga halo and other candidate halos.

As discussed in Manconi et al. (2024), possible two-zone diffusion scenarios and the proper motion of the Geminga pulsar do not affect the synchrotron signal from Geminga in the range of a few to tens of keV. These effects are more significant for older populations of e± emitting at GeV via inverse Compton scattering, rather than the X-ray and TeV-emitting populations, which have had the time to propagate away from the acceleration site. Given our interest in the pulsar’s halo diffuse emission, which is observed at TeV gamma rays to be extended a few degrees around the pulsar (corresponding to a physical size of a few tens of parsecs), the emission from the pulsar and its bow-shock nebula was not modeled here.

Using the benchmark model outlined above, we created an intensity map by integrating within the 4−12 keV energy band and varying the magnetic field in the interval 2 − 50 μG. These physical models represent the Mhalo sky maps, which are compared in the following sections with the one-day real-data observations, while in Section 5 they are used to estimate the sensitivity for a detection using longer observations. The resolution of the model sky map is 18″. Figure 1 shows an example for B = 3 μG. The color map represents the synchrotron flux from the Geminga pulsar halo, obtained with the benchmark model described above and integrated over the 4–12 keV energy range. The angular extension of the emission is compared with the XMM-Newton, SRG/ART-XC FOVs (green and white lines), as well as with the observed extension of the Geminga halo in gamma rays (dotted yellow line). By comparing the extensions in the sky of the Geminga halo synchrotron emission with the X-ray telescope FOV, the potential sensitivity of SRG/ART-XC to a significant portion of the source’s expected physical extent is evident.

thumbnail Fig. 1.

Model sky map Mhalo of the Geminga X-ray halo flux coming from synchrotron emission as integrated in the 4–12 keV band, for B = 3 μG. The field of view of the SRG/ART-XC observation is reported with a white square, and compared to the XMM-Newton field of view from Manconi et al. (2024) (green dashed circle), as well as with the extension of the TeV gamma rays observed by HAWC (dotted yellow circle).

A complementary view of the expected emission properties within the SRG/ART-XC observation window is presented in Figure 2, which illustrates the surface brightness, i.e., the synchrotron flux from the Geminga pulsar halo as a function of the distance from the center. Depending on the magnetic field value B, which is varied here from 2 μG to 6 μG, a decrease of a factor of approximately ten is predicted within the inner two degrees.

thumbnail Fig. 2.

Model prediction of the surface brightness of the Geminga X-ray halo flux from synchrotron emission, integrated over the 4−12 keV band within four degrees of the center. The magnetic field strength varies from 2 μG (dashed blue line) to 6 μG (dotted black line).

3. ART-XC observations of Geminga

In this section, we describe the observations of the Geminga region performed with ART-XC. We first detail the data collection strategy and then describe the data reduction methodology and the image fitting procedure.

3.1. Data collection

The region of Geminga was observed on 16 April 2023 with the Mikhail Pavlinsky Astronomical Roentgen Telescope – X-ray Concentrator (ART-XC, Pavlinsky et al. 2021), one of two X-ray telescopes on board the SRG observatory (Sunyaev et al. 2021), launched on 13 July 2019, from the Baikonur Cosmodrome.

The ART-XC instrument is an X-ray telescope assembly containing seven independent X-ray telescope modules. It covers the 4−30 keV energy range and extends the energy response of the other instrument on board SRG observatory, eROSITA (Predehl et al. 2021), which is sensitive in the 0.2−8 keV band.

The Geminga region was observed with a total exposure of 82 ksec1. in scanning mode, specially developed for X-ray imaging of extended objects with angular sizes greater than the FOV of the SRG instruments (Sunyaev et al. 2021; Krivonos et al. 2022; Semena et al. 2024). A single scan with a size of 3.5 × 3.5 degrees, centered on the Geminga pulsar position, was performed with a scan step of 4′. This step is optimized for the vignetting function of the ART-XC telescope to ensure a uniform exposure. As a result, the vignetting-corrected exposure reached ∼1800 seconds at each point.

3.2. Data analysis

We reduce the raw ART-XC telemetry data using the data analysis tools developed in IKI2, as outlined in Pavlinsky et al. (2021). Using the ARTPIPELINE software, we produced clean, calibrated event lists for each of the telescope modules and spacecraft attitude data. The cleaned science data were then processed with ARTPRODUCTS to obtain sky images, exposure, and particle background maps.

We combined the ART-XC cleaned event lists for all seven telescope modules into a single sky mosaic in the 4−12 keV energy band, which is the most sensitive energy range for detecting point-like and extended X-ray sources, given the energy dependence of the ART-XC effective area (Pavlinsky et al. 2016).

Figure 3 shows an adaptively smoothed sky image of Geminga obtained with the one-day ART-XC data in the 4−12 keV energy range. The data are presented in Galactic coordinates and in units of counts pix−1. The position of the Geminga pulsar is indicated by a white circle with a diameter of 1°. The point-like emission of the Geminga pulsar is not detected. The corresponding upper limit of 3 × 10−12 erg s−1 cm−2 (2σ) in the 4−12 keV band is consistent with a 4−12 keV flux of 4 × 10−13 erg s−1 cm−2 derived from a joint NuSTAR and XMM-Newton power-law fit above 3 keV, as reported by Mori et al. (2014).

thumbnail Fig. 3.

Sky image of the Geminga pulsar region in the 4−12 keV obtained with ART-XC, shown in units of counts pix−1. The image is smoothed with the dmimgadapt task from CIAO-4.15 using a Gaussian kernel. The corresponding exposure map is highly uniform, with ∼1800 seconds per pixel. The position of the Geminga pulsar is marked by a white circle of 1° diameter. The grid is displayed in Galactic coordinates. The green circle denotes the ART-XC FOV, 36′ in diameter.

In principle, the data contain uniformly distributed photons from the particle background and the cosmic X-ray background (CXB). In addition, a putative extended emission of Geminga may be present, which is expected to follow the physical halo model described in Section 2 (Mhalo, Fig. 1) and to be centered on the midpoint of the adopted scan. This pilot ART-XC observation provides a clean data set with uniform sky coverage, which we used to search for the Geminga halo and to guide the feasibility of future observations.

3.3. Image fitting procedure

We searched for X-ray emission from the Geminga pulsar halo using wide-field observations with ART-XC. Since the instrument operates in photon counting mode, we first converted the sky intensity map of the halo model to the ART-XC pixel frame. Next, we converted the model intensity (erg s−1 cm−2 pix−1) to counts (cts s−1 cm−2 pix−1), assuming an absorbed power-law model with photon index Γ = 2. We verified that this photon index adequately approximates the trend of the physical models over the considered energy range. The absorbing hydrogen column density of 3.57 × 1021 cm−2 was adopted from H14PI project (HI4PI Collaboration 2016).

To extract information on the putative contribution from the Geminga X-ray halo, we constructed a model for the number of counts within the observed region of interest (ROI). This model includes a uniform background, which comprises a combination of the particle background and CXB, and the Geminga halo model, Mhalo. We followed a widely used statistical paradigm, which considers the expected number of counts in each pixel, ⟨Npix⟩, predicted by the model, and observed counts, Npix, in the data map. The model incorporates two free parameters (Nbkg, Shalo), which are optimized on the data map. Specifically, we estimated the best-ft model parameters by maximizing the likelihood that the observed count map is described by the halo model. To do this, we used the SHERPA modeling and fitting package (Freeman et al. 2001; Siemiginowska et al. 2024), a part of the CIAO software (Fruscione et al. 2006), to construct a 2D model and fit it to the data.

As described above, our setup included a uniform background (comprising the particle background and CXB) and a space-dependent Geminga halo model, Mhalo, which was convolved with the ART-XC point spread function (PSF)3. The model setup within SHERPA is expressed as the sum of two contributions: psf(Mhalo)*const2d.Shalo*emap + const2d.Nbkg*emap. In this expression, psf(Mhalo) represents a convolution of a 2D sky map, Mhalo, with the PSF, which is normalized by the Shalo parameter and multiplied by the exposure map, emap. The background is added as const2d.Nbkg*emap. Technically, the fit parameters enter as const2d variables within the SHERPA fit. The background is not convolved with the PSF.

The two free parameters, Shalo and Nbkg , are expressed in units of cts s−1 pix−1. The fit was repeated for the synchrotron model, Mhalo, computed for different values of the magnetic field, B.

For each tested B, the fit results for the Shalo parameter can be interpreted as follows:

  1. Shalo ≈ 1 with a small error indicates a detection of the Geminga halo emission predicted by Mhalo for that fixed B.

  2. Shalo > 1 implies that the emission model must be amplified by a factor of Shalo.

  3. The non-detection of the Geminga halo is indicated by either a small value, Shalo ≈ 0, not consistent with unity, or by Shalo being consistent with both zero and unity within the errors. In the first case, the model predicts too many counts, while in the second case, we do not detect the halo emission and cannot constrain the model Mhalo.

The upper limit on the Geminga halo is obtained when Shalo ≈ 0 and is not consistent with unity. This case means that we do not detect the halo emission and that we can constrain the model Mhalo, and thus B. Based on the current pilot ART-XC data set, we realistically expected a non-detection, and the ability to place upper limits.

4. Results

Table 1 presents the best-fitting parameters of the background normalization, Nbkg, and scaling factor, Shalo, for the Geminga halo model, calculated for different test magnetic field strengths: 3, 10, 15, 30, and 50 μG. As seen from the table, the constant background is determined accurately; the values found for the different magnetic fields tested are consistent and within percent-level uncertainties. From the first line of the table, we see that the 3 μG Geminga halo is not detected, as expected given the current statistics of the ART-XC data set. The halo model with higher fields is also not detected. However, the fitting procedure can provide 1-σ upper limits.

Table 1.

Best-fitting parameters of the background normalization, Nbkg, and Geminga halo model scaling factor, Shalo, for different magnetic field strengths.

To identify the 1-σ upper limit on B, we calculated the model for additional field strengths below 10 μG. Figure 4 shows Shalo as a function of the magnetic field strength. According to the methodology in Section 3, the highest magnetic field consistent with the ART-XC dataset is 6 μG at the 68% confidence level. As shown in Fig. 4, the point for the Geminga halo model corresponding to 6 μG is consistent with zero and not with unity within errors. In other words, the limits derived from the one-day ART-XC scanning observation are not consistent with a magnetic field > 6 μG.

thumbnail Fig. 4.

Geminga halo emission model scaling factor, Shalo, obtained from one-day ART-XC data, shown as a function of magnetic field strength.

The X-ray flux upper limits derived from the one-day ART-XC observations are less constraining by a factor of three compared to existing limits obtained with narrow FOV telescopes (Manconi et al. 2024). We find it remarkable that, within just one-day observation, the physical parameters of the Geminga halo model can be constrained to values that are usually predicted by current state-of-the-art models of the Galactic magnetic field (Unger & Farrar 2024).

5. Simulations

To estimate the ability of ART-XC to improve the current upper limit on the detection of the Geminga halo emission, we performed extensive simulations covering longer observation times.

First, we simulated the 4−12 keV background image based on the expected background count rate. The ART-XC background at the L2 point slowly decreases with time, as a result of the current rise in solar activity. Fig. 5 shows the evolution of the ART-XC internal detector background in the 40−60 keV energy band, in which ART-XC does not have an energy response to incoming X-ray photons. However, since the one-day observation of Geminga in 2023, the background has decreased by ∼20%, along with increased solar flaring activity. Nevertheless, for the simulation of the 4−12 keV background map in 2025, we used 20% less than the 2023 background count rate, assuming a low probability of catching a flare during the observation. We then simulated the large-scale Geminga halo emission based on a given model setup. We combined the maps and ran the 2D fitting procedure described in Sect. 3.3, to estimate Shalo. By repeating the entire procedure 104 times, we obtained the distribution of Shalo values.

thumbnail Fig. 5.

Background count rate measured by ART-XCin the 60−120 keV band as a function of time in years. The vertical dashed line indicates the date of the SRG/ART-XC scanning observation of Geminga on 16 April 2023. The figure is based on data obtained from the SRG/ART-XC environment monitor, available at https://monitor.srg.cosmos.ru/.

Figure 6 shows the mean and 1-σ width derived from the distribution of Shalo values as a function of exposure time in days. We ran simulations for the Geminga halo model calculated for magnetic field strengths of 2, 3, 4, and 6 μG. The first value is relevant to our recent magnetic field constraint based on the XMM-Newton data (Manconi et al. 2024).

thumbnail Fig. 6.

Simulations of Geminga halo detection at different SRG/ART-XC exposures. The magnetic fields considered are 2, 3, 4, and 6 μG. The plot shows the Geminga emission model scaling factor, Shalo, and its 1-σ uncertainty. The dotted red line indicates Shalo = 1.

Simulation of all the explored values of the magnetic field confirms the results of the current work for the one-day Geminga observation in 2023, showing no detection of the halo and an upper limit for one-day observation at around 6 μG. As seen in the figure, the ten-day exposure should improve the one-day upper limit by a factor of two. Increasing the exposure time to 20 days would result in the detection of the Geminga halo emission corresponding to the model prediction for B = 3 μG at the 68% probability level. Finally, we note that the Geminga observation mode can be changed from scan to grid tiling with smaller sky coverage, e.g., 1.5° ×1.5°, which is better optimized for a 2−3 μG magnetic field (Fig. 2) and requires less exposure time than scan mode.

6. Summary and conclusion

In this work, we performed a wide-field search for the putative large-scale X-ray halo around the Geminga pulsar using a scanning, one-day observation with the ART-XC telescope onboard the SRG observatory. This probe observation covers a large area of the sky, comparable with the size of expected Geminga X-ray emission. Thus, compared to other X-ray searches for the Geminga halo, the ART-XC provides a unique data set characterized by direct imaging in the 4−12 keV energy band and a highly uniform sky coverage. We developed a physical model scaled on the existing gamma-ray measurements of the Geminga pulsar halo at high energies (GeV/TeV), which predicts spatially distributed synchrotron X-ray halo flux. This model predicts the X-ray flux in the 4−12 keV energy band as a function of magnetic field strength. We performed an image-fitting procedure in which the normalization of the intensity of the Geminga synchrotron halo emission and the uniform background level enter the fit as free parameters. We find that the model with a magnetic field of 6 μG is compatible with the ART-XC data at 68% confidence and therefore provides an upper limit for the magnetic field strength. The moderate constraint on the magnetic field, higher by about a factor of two compared to other state-of-the-art studies, is mainly governed by the high instrumental background. Remarkably, within just a one-day ART-XC observation, the magnetic field around the Geminga halo is constrained to values of the same order of magnitude as the expected Galactic magnetic field strength. Additionally, using extensive simulations, we show that longer ART-XC observations of Geminga can significantly improve these constraints. A ten-day exposure can improve the upper limit on the magnetic field by a factor of two relative to the one-day observation. Observations of about 20 days or more have the potential to uncover the synchrotron emission from the Geminga pulsar halo.

The techniques developed in this work for ART-XC can be generalized to test other theoretical models of the Geminga pulsar halo, in addition to the benchmark suppressed diffusion model investigated here. In such cases, it would be necessary to build a consistent spectral and spatial template. Furthermore, the techniques developed and applied to the ART-XC data in this study are applicable to data from other wide field-of-view X-ray instruments, such as the Advanced X-ray Imaging Satellite (AXIS; Reynolds et al. 2023). Detecting an X-ray counterpart of the Geminga gamma-ray halo would provide a unique probe of particle acceleration and electron-positron transport around pulsars. Such a detection would inform current theoretical models, for example, regarding the structure of the magnetic field through which these relativistic particles propagate, and improve our understanding of the halo phenomenon around Galactic pulsars.

Data availability

Figure 3 (FITS files) is available at the CDS via https://cdsarc.cds.unistra.fr/viz-bin/cat/J/A+A/705/A107. At the time of writing, the SRG/ART-XC data and the corresponding data analysis software have a private status. Public access to the ART-XC scientific archive will be provided in the future.

Acknowledgments

The Mikhail Pavlinsky ART-XC telescope is the hard X-ray instrument on board the SRG observatory, a flagship astrophysical project of the Russian Federal Space Program realized by the Russian Space Agency in the interests of the Russian Academy of Sciences. The ART-XC team thanks the Russian Space Agency, Russian Academy of Sciences, and State Corporation Rosatom for the support of the SRG project and ART-XC telescope. SM acknowledges the European Union’s Horizon Europe research and innovation program for support under the Marie Sklodowska-Curie Action HE MSCA PF–2021, grant agreement No.10106280, project VerSi and the support of the French Agence Nationale de la Recherche (ANR), under the grant ANR-24-CPJ1-0121-01. F.D. and M.D.M. acknowledge support from the Research grant TAsP (Theoretical Astroparticle Physics) funded by Istituto Nazionale di Fisica Nucleare (INFN). M.D.M. acknowledges support from the Italian Ministry of University and Research (MUR), PRIN 2022 “EXSKALIBUR – Euclid-Cross-SKA: Likelihood Inference Building for Universe’s Research”, Grant No. 20222BBYB9, CUP I53D23000610 0006, and from the European Union – Next Generation EU.

References

  1. Abdo, A. A., Allen, B. T., Aune, T., et al. 2009, ApJ, 700, L127 [CrossRef] [Google Scholar]
  2. Abeysekara, A. U., Albert, A., Alfaro, R., et al. 2017, Science, 358, 911 [NASA ADS] [CrossRef] [Google Scholar]
  3. Adams, C. B., Archer, A., Bangale, P., et al. 2025, ApJ, 985, 90 [Google Scholar]
  4. Aharonian, F., An Axikegu, Q., Bai, L. X., et al. 2021, Phys. Rev. Lett., 126, 241103 [NASA ADS] [CrossRef] [Google Scholar]
  5. Aharonian, F., Aschersleben, J., Backes, M., et al. 2023, A&A, 673, A148 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Amato, E., & Recchia, S. 2024, Riv. Nuovo Cim., 47, 399 [Google Scholar]
  7. Di Mauro, M., Manconi, S., & Donato, F. 2019, Phys. Rev. D, 100, 123015 [NASA ADS] [CrossRef] [Google Scholar]
  8. Evoli, C., Amato, E., Blasi, P., & Aloisio, R. 2021, Phys. Rev. D, 103, 083010 [NASA ADS] [CrossRef] [Google Scholar]
  9. Fang, K. 2022, Front. Astron. Space Sci., 9, 1022100 [NASA ADS] [CrossRef] [Google Scholar]
  10. Freeman, P., Doe, S., & Siemiginowska, A. 2001, in Astronomical Data Analysis, eds. J. L. Starck, & F. D. Murtagh, SPIE Conf. Ser., 4477, 76 [Google Scholar]
  11. Fruscione, A., McDowell, J. C., Allen, G. E., et al. 2006, in Observatory Operations: Strategies, Processes, and Systems, eds. D. R. Silva, & R. E. Doxsey, SPIE Conf. Ser., 6270, 62701V [NASA ADS] [CrossRef] [Google Scholar]
  12. Gaensler, B. M., & Slane, P. O. 2006, ARA&A, 44, 17 [Google Scholar]
  13. HI4PI Collaboration (Ben Bekhti, N., et al.) 2016, A&A, 594, A116 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  14. Khokhriakova, A., Becker, W., Ponti, G., et al. 2024, A&A, 683, A180 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  15. Khokhriakova, A., Becker, W., Predehl, P., et al. 2025, Res. Notes AAS, 9, 154 [Google Scholar]
  16. Krivonos, R., Arefiev, V., Lapshov, I., et al. 2022, MNRAS, 510, 3113 [NASA ADS] [CrossRef] [Google Scholar]
  17. Krivonos, R., Burenin, R., Filippova, E., et al. 2025, Exp. Astron., 59, 37 [Google Scholar]
  18. Liu, R.-Y. 2022, Int. J. Mod. Phys. A, 37, 2230011 [NASA ADS] [CrossRef] [Google Scholar]
  19. Liu, R.-Y., Ge, C., Sun, X.-N., & Wang, X.-Y. 2019, ApJ, 875, 149 [NASA ADS] [CrossRef] [Google Scholar]
  20. López-Coto, R., de Oña Wilhelmi, E., Aharonian, F., Amato, E., & Hinton, J. 2022, Nat. Astron., 6, 199 [CrossRef] [Google Scholar]
  21. Manconi, S., Di Mauro, M., & Donato, F. 2020, Phys. Rev. D, 102, 023015 [NASA ADS] [CrossRef] [Google Scholar]
  22. Manconi, S., Woo, J., Shang, R.-Y., et al. 2024, A&A, 689, A326 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  23. Mori, K., Wu, Y. W., Immer, K., et al. 2014, ApJ, 793, 88 [NASA ADS] [CrossRef] [Google Scholar]
  24. Niu, S., Yuan, Q., Zhang, S. N., et al. 2025, ArXiv e-prints [arXiv:2501.17046] [Google Scholar]
  25. Orusa, L., Manconi, S., Donato, F., & Di Mauro, M. 2025, JCAP, 02, 029 [Google Scholar]
  26. Pavlinsky, M., Akimov, V., Levin, V., et al. 2016, in Status of ART-XC/SRG instrument, SPIE Conf. Ser., 9905, 99051J [Google Scholar]
  27. Pavlinsky, M., Tkachenko, A., Levin, V., et al. 2021, A&A, 650, A42 [EDP Sciences] [Google Scholar]
  28. Philippov, A., & Kramer, M. 2022, ARA&A, 60, 495 [NASA ADS] [CrossRef] [Google Scholar]
  29. Plucinsky, P. P., Snowden, S. L., Aschenbach, B., et al. 1996, ApJ, 463, 224 [CrossRef] [Google Scholar]
  30. Posselt, B., Pavlov, G. G., Slane, P. O., et al. 2017, ApJ, 835, 66 [CrossRef] [Google Scholar]
  31. Predehl, P., Andritschke, R., Arefiev, V., et al. 2021, A&A, 647, A1 [EDP Sciences] [Google Scholar]
  32. Reynolds, C. S., Kara, E. A., Mushotzky, R. F., et al. 2023, Proc. SPIE Int. Soc. Opt. Eng., 12678, 126781E [Google Scholar]
  33. Semena, A., Mereminskiy, I., Lutovinov, A., et al. 2024, MNRAS, 529, 941 [NASA ADS] [CrossRef] [Google Scholar]
  34. Siemiginowska, A., Burke, D., Günther, H. M., et al. 2024, ApJS, 274, 43 [Google Scholar]
  35. Sunyaev, R., Arefiev, V., Babyshkin, V., et al. 2021, A&A, 656, A132 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  36. Unger, M., & Farrar, G. R. 2024, ApJ, 970, 95 [NASA ADS] [CrossRef] [Google Scholar]
  37. Vernetto, S., & Lipari, P. 2016, Phys. Rev. D, 94, 063009 [NASA ADS] [CrossRef] [Google Scholar]
  38. Wu, Q.-Z., Li, C.-M., Liang, X.-H., Ge, C., & Liu, R.-Y. 2024, ApJ, 969, 9 [Google Scholar]
1

The eROSITA telescope was not operating during this period.

2

Space Research Institute of the Russian Academy of Sciences, Moscow, Russia.

3

Given the large angular size of the Geminga pulsar halo, the knowledge of the ART-XC mirror’s response at large radii is crucial. We used the updated in-flight PSF calibrations made by Krivonos et al. (2025), who used the observations of the bright X-ray source Sco X-1 to calibrate the ART-XC PSF shape up to ∼1.5°.

All Tables

Table 1.

Best-fitting parameters of the background normalization, Nbkg, and Geminga halo model scaling factor, Shalo, for different magnetic field strengths.

In the text

All Figures

thumbnail Fig. 1.

Model sky map Mhalo of the Geminga X-ray halo flux coming from synchrotron emission as integrated in the 4–12 keV band, for B = 3 μG. The field of view of the SRG/ART-XC observation is reported with a white square, and compared to the XMM-Newton field of view from Manconi et al. (2024) (green dashed circle), as well as with the extension of the TeV gamma rays observed by HAWC (dotted yellow circle).
In the text
thumbnail Fig. 2.

Model prediction of the surface brightness of the Geminga X-ray halo flux from synchrotron emission, integrated over the 4−12 keV band within four degrees of the center. The magnetic field strength varies from 2 μG (dashed blue line) to 6 μG (dotted black line).
In the text
thumbnail Fig. 3.

Sky image of the Geminga pulsar region in the 4−12 keV obtained with ART-XC, shown in units of counts pix−1. The image is smoothed with the dmimgadapt task from CIAO-4.15 using a Gaussian kernel. The corresponding exposure map is highly uniform, with ∼1800 seconds per pixel. The position of the Geminga pulsar is marked by a white circle of 1° diameter. The grid is displayed in Galactic coordinates. The green circle denotes the ART-XC FOV, 36′ in diameter.
In the text
thumbnail Fig. 4.

Geminga halo emission model scaling factor, Shalo, obtained from one-day ART-XC data, shown as a function of magnetic field strength.
In the text
thumbnail Fig. 5.

Background count rate measured by ART-XCin the 60−120 keV band as a function of time in years. The vertical dashed line indicates the date of the SRG/ART-XC scanning observation of Geminga on 16 April 2023. The figure is based on data obtained from the SRG/ART-XC environment monitor, available at https://monitor.srg.cosmos.ru/.
In the text
thumbnail Fig. 6.

Simulations of Geminga halo detection at different SRG/ART-XC exposures. The magnetic fields considered are 2, 3, 4, and 6 μG. The plot shows the Geminga emission model scaling factor, Shalo, and its 1-σ uncertainty. The dotted red line indicates Shalo = 1.
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.