仪器与数据

Instrument and data

simulation结果不是望远镜直接观测。pyXSIM按gas temperature/metallicity/density以CIE APEC Monte Carlo生成2 Mpc内photons与LOS Doppler shifts；SOXS再用SIXTE提供的PSF/ARF/RMF分别模拟eROSITA七个mirror assemblies并stack。使用z=0.1 simulation snapshots，每个synthetic observation从对应observed sample随机赋redshift并采用1000 ks；因比较对象已cleaned/background-subtracted，不加foreground/background。

The simulation results are not direct telescope observations. pyXSIM Monte Carlo samples CIE APEC photons within 2 Mpc from gas temperature, metallicity, and density and applies line-of-sight Doppler shifts. SOXS then uses SIXTE PSF/ARF/RMF files to simulate and stack the seven eROSITA mirror assemblies. The analysis uses z=0.1 simulation snapshots, assigns each synthetic observation a redshift drawn from the corresponding observed sample, and adopts 1000 ks; no foreground/background is added because the comparator is already cleaned and background-subtracted.

观测视线与样本域

Sightlines and sample domain

每个simulation galaxy沿x、y、z三方向生成external-observer LOS projections后再stack，不是一条真实M31 LOS。Figure 3采用11<log(M*/Msun)<11.25 bin和projected physical 10-30 kpc annulus；论文没有把这些synthetic LOS重新放到M31 sky coordinates或穿过MW absorption。

Each simulated galaxy is projected along the x, y, and z directions and stacked as external-observer sightlines; none is a real M31 sightline. Figure 3 adopts the 11<log(M*/Msun)<11.25 bin and projected physical 10-30 kpc annulus. The paper does not place these synthetic sightlines at M31 sky coordinates or pass them through Milky-Way absorption.

原文测量量

Native measurement

native plotted quantity是simulation的intrinsic X-ray luminosity surface density，Figure 7给各model line与spread。Figure 7意图与Zhang的0.5-2.0 keV profile比较，邻近Figure 6也把soft X-ray定义为0.5-2.0 keV；但profile methods/caption没有显式闭合Figure 7的event-energy cut或frame。当前数值由author arXiv figure在校准log axes和RGB后digitize。

The native plotted quantity is simulated intrinsic X-ray luminosity surface density, with Figure 7 showing model lines and spreads. Figure 7 is intended to match the Zhang 0.5-2.0 keV profiles, and adjacent Figure 6 defines soft X-rays as 0.5-2.0 keV, but the profile methods/caption do not explicitly close the Figure 7 event-energy cut or frame. Current values are digitized from the author arXiv figure using calibrated log axes and exact RGB values.

原文模型

Published model

EAGLE Ref-L050N0752用single-mode thermal AGN feedback、DeltaT_AGN=10^8.5 K；EAGLE-AGNdT9提高到10^9 K并改变accretion parameterization；EAGLE-NoAGN关闭AGN feedback。SIMBA标准模型包含high-Eddington kinetic wind、low-Eddington jet及相关X-ray mode；SIMBA-NoAGN全部关闭。五者是不同subgrid physics predictions，不是同一measurement的statistical error models。

EAGLE Ref-L050N0752 uses single-mode thermal AGN feedback with DeltaT_AGN=10^8.5 K; EAGLE-AGNdT9 raises this to 10^9 K and modifies the accretion parameterization; EAGLE-NoAGN disables AGN feedback. Fiducial SIMBA includes a high-Eddington kinetic wind, a low-Eddington jet, and associated X-ray mode; SIMBA-NoAGN disables AGN feedback. These are distinct subgrid-physics predictions, not statistical-error models for one measurement.

质量scaling与radial profile

Mass scaling and radial profile

原文没有给一条可直接把任意galaxy mass连续缩放到M31的analytic S_X-M relation。Figure 7按stellar mass分bin后显示：NoAGN在各mass bins的inner profile通常更亮；log M*<11.25时inner CGM的SIMBA显著亮于EAGLE，而最高mass时两套fiducial profiles更接近。作者先在0.1 dex stellar-mass cells重采样simulation，使其匹配Zhang observed stellar-mass distribution；本项目随后直接选择11<log(M*/Msun)<11.25、同时限制12<log(M200c/Msun)<14的M31-mass panel。halo-mass distribution没有与观测强制匹配。radius也不做R/R200重标度：Figure 3读取原文physical 10-30 kpc bin。

The paper does not provide an analytic S_X-mass relation that continuously rescales an arbitrary galaxy to M31. Across the stellar-mass bins in Figure 7, NoAGN runs are generally brighter in the inner profile; below log M*=11.25, SIMBA is substantially brighter than EAGLE in the inner CGM, while the two fiducial profiles converge at the highest masses. The paper resamples each simulation in 0.1 dex stellar-mass cells to match the observed Zhang stellar-mass distribution; this project then directly selects the M31-mass panel, 11<log(M*/Msun)<11.25 with 12<log(M200c/Msun)<14. The halo-mass distribution is not forced to match the observations. Radius is not rescaled by R/R200 either: Figure 3 reads the published physical 10-30 kpc bin.

原文结果

Published results

Figure 7显示在M31-mass bin与10-30 kpc内，NoAGN尤其SIMBA-NoAGN显著提高soft-X-ray brightness；不同feedback prescriptions给出数量级差异。

Figure 7 shows that within the M31-mass bin and 10-30 kpc, NoAGN models, especially SIMBA-NoAGN, strongly increase soft-X-ray brightness; different feedback prescriptions span orders of magnitude.

如何进入M31 Figure 3

Transfer into M31 Figure 3

每个model从Figure 7 digitize 10-30 kpc intrinsic luminosity-density central/spread；使用同一distance-cancel surface-brightness conversion和v19 APEC absorption bridge得到Figure 3 values。五个points保持离散，不插值成M31 radial curves。

For each model, the central value and spread of intrinsic luminosity density in 10-30 kpc are digitized from Figure 7. The same distance-cancel surface-brightness conversion and v19 APEC absorption bridge give the Figure 3 values. The five points remain discrete and are never interpolated into M31 radial curves.

假设与解释边界

Assumptions and boundaries

转换对所有simulation采用单一v19 APEC spectrum，而非各模拟自身multiphase spectra。Figure 7范围是non-resampled galaxies的digitized bootstrap envelope，但论文没有说明confidence level，不能称为95% confidence interval。synthetic profiles已在随机赋予的观测redshifts下经过eROSITA PSF/response；单位转换不会把它们变成M31/XMM空间响应。stellar-mass matching不保证M31 halo mass、SFR、morphology或environment完全匹配。论文明确指出两个NoAGN runs不能再现galaxy stellar-mass function或stellar-halo mass relation，因此只能用作feedback sensitivity experiments。

The transfer applies one v19 APEC spectrum to all simulations rather than each simulation's multiphase spectrum. The Figure 7 ranges are digitized bootstrap envelopes for the non-resampled galaxies, but the confidence level is not stated and they are not a 95% confidence interval. The synthetic profiles are already convolved with the eROSITA PSF/response at randomly assigned observed redshifts; a unit conversion does not turn them into an M31/XMM spatial response. Stellar-mass matching does not guarantee matched M31 halo mass, SFR, morphology, or environment. The paper explicitly notes that the two NoAGN runs fail to reproduce the galaxy stellar-mass function or stellar-to-halo mass relation, so they are feedback-sensitivity experiments rather than realistic universes.

本论文对应的Figure 3数据点

Figure 3 points from this study

Primary-source证据地图

Primary-source evidence map

• publisher
• pdf
• arxiv