X‑SRG: SRG X‑Ray Survey Mission

• X‑SRG is a comprehensive X‑ray survey system using co‑aligned eROSITA and ART‑XC telescopes to cover 0.2–30 keV.
• It employs advanced grazing‑incidence optics and repeated all‑sky scans to detect galaxy clusters, AGN, and Galactic sources with high precision.
• Its survey design, with deep and repeated sky maps, enables time‑domain studies and cross‑wavelength calibration of diffuse and extended X‑ray structures.

X‑SRG denotes the X‑ray survey capability of the Spectrum‑Roentgen‑Gamma observatory, a Russian–German mission built around two co‑aligned grazing‑incidence telescopes: eROSITA in the soft and medium X‑ray band and the Mikhail Pavlinsky ART‑XC telescope at higher energies. In the literature, the term is used to refer either specifically to the SRG/eROSITA all‑sky X‑ray survey or more broadly to the combined X‑ray program of SRG, whose original design emphasis was cosmology through large cluster samples and whose realized flight program extends to active galactic nuclei, Galactic compact objects, diffuse emission, and time‑domain X‑ray astrophysics (Cappelluti et al., 2010, Predehl et al., 2020, Sunyaev et al., 2021).

1. Mission definition and historical development

SRG was conceived as a dedicated all‑sky X‑ray survey mission followed by pointed observations. The early mission description placed the observatory on the Russian “Navigator” satellite, in orbit around the Sun–Earth L2 point, with eROSITA as the core soft‑X‑ray instrument and ART‑XC as the complementary harder‑band instrument; at that stage, eROSITA was in phase C/D and the launch was scheduled for late 2012 (Cappelluti et al., 2010). The flown observatory was subsequently launched on 13 July 2019 from Baikonur and inserted into a large‑amplitude halo orbit around the Sun–Earth L2 point, where the stable thermal and particle environment is central to its survey performance (Predehl et al., 2020).

The operational program is defined by repeated full‑sky scans. After commissioning and calibration/performance verification, eROSITA started surveying the entire sky on 13 December 2019, with eight complete scans planned over four years, each lasting about six months; later mission documents describe a post‑survey phase of pointed and scanning observations via open calls (Predehl et al., 2020). A mission‑level overview written during survey operations states that by mid‑June 2021 the third six‑month all‑sky survey had been completed and emphasizes the scientific logic of eight independent sky maps: co‑addition for depth and repeated visits for variability studies on six‑month timescales (Sunyaev et al., 2021).

Within this architecture, “X‑SRG” is not a single instrument but a coordinated broad‑band survey system. eROSITA provides the primary soft‑band survey, source localization, and large collecting area below a few keV, while ART‑XC extends the observatory into the 4–30 keV regime, where obscured active nuclei, hard Galactic sources, and hotter thermal plasmas become accessible (Pavlinsky et al., 2021).

2. Observatory architecture and instrumental characteristics

eROSITA is a seven‑telescope Wolter‑I array. Each mirror assembly contains 54 nested shells with outer diameter about 360 mm, and each telescope has its own pn‑CCD camera with a $384\times384$ pixel imaging area of $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$. The focal length is 1600 mm, the pixel scale is $9.6^{\prime\prime}\times9.6^{\prime\prime}$, the field of view is approximately $1.03^\circ$ in diameter, and the nominal integration time is 50 ms. The CCDs are passively cooled to about $-85^\circ\mathrm{C}$, and each camera includes a four‑position filter wheel with calibration sources providing Al K$\alpha$, Ti K$\alpha$, and Mn K$\alpha$ lines (Predehl et al., 2020).

The eROSITA angular response was deliberately optimized for survey work rather than for the narrowest possible on‑axis image core. Early design papers quoted an on‑axis PSF half‑energy width of about $15^{\prime\prime}$ and a practical survey‑mode resolution of $25^{\prime\prime}$–$28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$0 (Cappelluti et al., 2010). The flight instrument description reports that the cameras were placed 0.4 mm intrafocal, slightly worsening the on‑axis HEW to about $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$1 while improving the field‑averaged HEW to about $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$2, which is the more relevant quantity for scanning survey performance (Predehl et al., 2020). In collecting power, the early concept quoted $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$3 at 1.5 keV (Cappelluti et al., 2010), whereas the in‑flight instrument paper gives a total on‑axis effective area of roughly $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$4 at 1.5 keV and a field‑averaged soft‑band effective area of $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$5 at 1 keV (Predehl et al., 2020). The same paper states that the grasp $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$6 is the largest of any imaging soft X‑ray telescope in approximately 0.3–3.5 keV (Predehl et al., 2020).

ART‑XC is the hard‑X‑ray component of X‑SRG. It consists of seven co‑aligned Wolter‑I mirror systems, each with 28 nested Ni/Co shells coated with iridium, feeding CdTe double‑sided strip detectors. The nominal energy range is 4–30 keV, with the all‑sky survey optimized for 4–12 keV. The focal length is 2700 mm, the detectors provide $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$7 imaging pixels with angular size about $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$8, and the timing resolution is $28.8\ \mathrm{mm}\times28.8\ \mathrm{mm}$9 (Pavlinsky et al., 2021). The total on‑axis effective area is about $9.6^{\prime\prime}\times9.6^{\prime\prime}$0 at 8.1 keV (Pavlinsky et al., 2021), while a later pulsar‑discovery paper describes the telescope as having total effective area $9.6^{\prime\prime}\times9.6^{\prime\prime}$1 at 6 keV and a PSF of about $9.6^{\prime\prime}\times9.6^{\prime\prime}$2 (Lutovinov et al., 2021). In survey mode, the effective point‑spread function has FWHM $9.6^{\prime\prime}\times9.6^{\prime\prime}$3, and in‑flight performance has been reported as very close to pre‑launch expectations (Pavlinsky et al., 2021).

Taken together, eROSITA and ART‑XC give SRG continuous X‑ray coverage from 0.2 to 30 keV, combining soft‑band survey grasp with hard‑band imaging and timing (Pavlinsky et al., 2021).

3. Survey strategy, cadence, and sensitivity

The survey geometry is set by spacecraft rotation. In eROSITA survey mode, the spacecraft rotates continuously with a scan rate of $9.6^{\prime\prime}\times9.6^{\prime\prime}$4, completing one full rotation every 4 h; a source crosses the central field of view in about 40 s, and locations near a scan path are revisited roughly six times per day (Predehl et al., 2020). The rotation axis approximately follows the Sun, so a full‑sky survey requires about six months. The same general architecture applies to SRG as a whole, and mission summaries emphasize that the repeated scans generate both an all‑sky depth map and a built‑in variability experiment on timescales from hours to years (Sunyaev et al., 2021).

Exposure is strongly latitude dependent. The first eROSITA all‑sky survey yielded unvignetted exposures of about 200 s near the ecliptic equator and more than 10,000 s near the ecliptic poles; for the full four‑year survey, the instrument paper quotes $9.6^{\prime\prime}\times9.6^{\prime\prime}$5 s at the ecliptic plane, a sky average of $9.6^{\prime\prime}\times9.6^{\prime\prime}$6 s, and $9.6^{\prime\prime}\times9.6^{\prime\prime}$7 s at the poles (Predehl et al., 2020). Earlier mission planning described the same strategy in survey‑science terms as an average exposure of roughly 3 ks per sky position and deep polar regions of 20–40 ks (Cappelluti et al., 2010).

Sensitivity scales accordingly. For the completed eight‑scan eROSITA survey, predicted point‑source limits are $9.6^{\prime\prime}\times9.6^{\prime\prime}$8 in 0.2–2.3 keV and $9.6^{\prime\prime}\times9.6^{\prime\prime}$9 in 2.3–8 keV near the ecliptic equator, improving to $1.03^\circ$0 and $1.03^\circ$1, respectively, near the poles (Predehl et al., 2020). The soft‑band survey is about 25 times deeper than the ROSAT All‑Sky Survey, while the 2.3–8 keV program constitutes the first true imaging all‑sky survey in that band (Predehl et al., 2020). For ART‑XC, the four‑year mission is described as the first true imaging all‑sky survey with grazing‑incidence optics in 4–30 keV and the deepest and sharpest all‑sky map in 4–12 keV; the final catalogue is expected to contain about 5000 sources (Pavlinsky et al., 2021).

The survey design also determines the time domain. SRG mission summaries stress that the eight sky maps enable monitoring of long‑term variability every six months for a huge number of Galactic and extragalactic sources, while the 4‑h spacecraft rotation yields intra‑day sampling for bright sources during each survey passage (Sunyaev et al., 2021).

4. Extragalactic science: cosmology, AGN censuses, and source identification

The design‑driving science of X‑SRG is cosmology through galaxy clusters. The original eROSITA mission paper specified the detection of about 50,000–100,000 clusters of galaxies out to $1.03^\circ$2, with essentially all evolved clusters above $1.03^\circ$3 detectable up to $1.03^\circ$4. The intended cosmological observables were the cluster mass function $1.03^\circ$5, its evolution $1.03^\circ$6, the cluster power spectrum $1.03^\circ$7, baryonic acoustic features, the cluster baryon fraction, and X‑ray plus Sunyaev–Zeldovich distance tests (Cappelluti et al., 2010). The flown mission overview later reformulated this as a survey expected to detect $1.03^\circ$8 clusters and a few million AGN, with the cluster sample spanning the regime needed to constrain $1.03^\circ$9, $-85^\circ\mathrm{C}$0, and dark‑energy parameters (Predehl et al., 2020).

The AGN program is equally central. Pre‑mission estimates in the design papers extrapolated $-85^\circ\mathrm{C}$1 relations to predict $-85^\circ\mathrm{C}$2 to $-85^\circ\mathrm{C}$3 AGN detections, extending to redshifts $-85^\circ\mathrm{C}$4–8 (Cappelluti et al., 2010). During operations, eRASS:1 alone detected 1,004,624 X‑ray sources across both hemispheres, illustrating the scale of the survey catalogues (Predehl et al., 2020). Hard‑band selection through ART‑XC adds a less absorption‑biased view of the nearby AGN population: a study of 14 sources selected in 4–12 keV from the first five all‑sky surveys found that all were Seyfert galaxies at $-85^\circ\mathrm{C}$5–0.238, six of them first detected in X‑rays by SRG, with four showing intrinsic absorption above $-85^\circ\mathrm{C}$6 at 90% confidence and one probably heavily obscured at $-85^\circ\mathrm{C}$7 (Uskov et al., 2023). That same identification program projects a statistically complete hard‑band AGN sample of $-85^\circ\mathrm{C}$8 objects by the end of the planned eight surveys (Uskov et al., 2023).

Because X‑SRG source densities far exceed what can be exploited without systematic counterpart identification, the mission has also driven methodological work on catalogue construction. In the Lockman Hole, a neural‑network‑assisted probabilistic optical cross‑match with DESI Legacy Imaging Surveys data reached 94% precision for the entire X‑ray catalogue and 97% for sources with $-85^\circ\mathrm{C}$9, providing a template for large‑scale counterpart association in the all‑sky survey (Bykov et al., 2023). A plausible implication is that X‑SRG should be understood not only as a telescope system but also as a survey infrastructure in which astrophysical yield depends on coordinated X‑ray, optical, and infrared cataloguing.

5. Galactic compact objects and time‑domain discoveries

Repeated all‑sky passes and the complementarity of eROSITA and ART‑XC make X‑SRG particularly effective for faint, persistent, and moderately variable Galactic sources that fall between classical soft monitors and bright hard‑X‑ray transients. This is illustrated by SRGA J204318.2+443815 / SRGe J204319.0+443820, discovered in the second all‑sky survey: ART‑XC found the hard source, eROSITA supplied the precise position, and follow‑up with XMM‑Newton, NICER, and NuSTAR established a coherent period of about 742 s, a pulsed fraction increasing from about 20% in soft X‑rays to $\alpha$0 at higher energies, a hard spectrum with exponential cutoff, and a characteristic luminosity $\alpha$1, leading to classification as a persistent, low‑luminosity Be/X‑ray binary pulsar with a B0–B2e companion (Lutovinov et al., 2021).

A closely related case is SRGA J124404.1−632232 / SRGU J124403.8−632231, detected as a new source in the third SRG survey. NuSTAR timing established $\alpha$2 s, SALT spectroscopy revealed a heavily reddened Be star with strong H$\alpha$3 emission, and broadband modelling placed the source in the low‑luminosity Be/XRB regime with $\alpha$4 (Doroshenko et al., 2021). In the Magellanic Clouds, SRG/eROSITA detections likewise triggered XMM‑Newton follow‑up that uncovered pulsations at 40.6 s, 17.3 s, and 784 s in three LMC Be/X‑ray binaries, again showing the survey‑plus‑follow‑up logic of the mission (Haberl et al., 2023).

SRG has also been used to characterize known but unusual pulsars. For SXP 1323 in the Small Magellanic Cloud, combined ART‑XC and eROSITA data produced the first broadband 1–20 keV spectrum and a measurement of pulsed fraction up to 16 keV, while addition of archival XMM‑Newton timing showed that after 2016 the source entered a linear spin‑up phase with $\alpha$5 (Mereminskiy et al., 2021). In Her X‑1, eROSITA observed two low states during the all‑sky survey and detected orbital modulation when the neutron star was hidden from direct view, supporting an interpretation in which the observed flux is scattered by a hot corona above the optical star, a hot halo above the accretion disk, and the optically thick cold atmosphere of the donor (Shakura et al., 2021).

These results support the claim, made explicitly in SRG pulsar papers, that the observatory can unveil a hidden population of faint persistent objects, including slowly rotating Be‑system pulsars (Lutovinov et al., 2021).

6. Diffuse emission, extended structures, and foreground calibration

X‑SRG has a distinct role in diffuse and extended X‑ray astrophysics because eROSITA operates at L2, where the magnetosheath contribution to solar‑wind charge‑exchange emission is expected to be negligible. In a shadowing study of three giant molecular clouds, eRASS1–4 data were used to separate heliospheric SWCX from the Local Hot Bubble and background halo emission. The derived LHB electron density was $\alpha$6 and approximately independent of sight line, while the temperature varied from $\alpha$7 keV toward Chamaeleon II and III to $\alpha$8 keV toward Ophiuchus and $\alpha$9 keV toward Corona Australis; the SWCX contribution increased monotonically from the start of 2020 and dominated the 0.3–0.7 keV foreground toward Ophiuchus from eRASS2 onward (Yeung et al., 2023). A later western‑hemisphere analysis went further by constructing a “dark sky” map with residual heliospheric contamination of about 1 LU and using the remaining SWCX signal to map the flow of interstellar matter through the Solar System (Dennerl et al., 28 Apr 2026).

For Galactic extended sources, eROSITA and ART‑XC probe different physical regimes. Spatially resolved eROSITA spectroscopy of the SNR G18.95−1.1 found that the X‑ray‑dim northern regions can be described by thin thermal plasma with temperature about 0.3 keV and solar composition, while bright southern clumps along the ridge are consistent with Si‑rich ejecta at about 0.3 keV, with abundance ratios suggesting origin in deeper layers of the progenitor star (Bykov et al., 2021). ART‑XC, operating in the harder band, mapped Puppis A above 4 keV and detected significant extended emission in 4–6 keV with shell‑rim morphology, an extended hard component, a bright northeastern knot, and four point sources, three previously catalogued and one newly discovered (Krivonos et al., 2021).

These studies show that X‑SRG is not limited to source counting. It also functions as a calibration platform for the soft diffuse foreground, a survey instrument for large‑scale hot plasmas, and a wide‑field imager for extended Galactic remnants. This suggests that the long‑term significance of X‑SRG lies as much in the standardization of the X‑ray sky—through repeated full‑sky imaging, foreground decomposition, and uniform source catalogues—as in any single headline sample.