Abstract

We consider the population of point X-ray sources from the two-year SRG/eROSITA survey in the 1 \(\%\) DESI spectroscopic survey region in the eastern Galactic hemisphere (eROSITA–1 \(\%\) DESI–East). The data under consideration combine a large survey area (compared to the medium-format Stripe82X and XMM-XXL X-ray surveys) and a record high completeness of spectroscopy for the optical counterparts of X-ray sources. We compare the results of the photometric (SRGz) and spectroscopic/astrometric (DESI EDR, SDSS, HELP, GAIA) measurements of the classes and redshifts of objects in three X-ray-flux-limited eROSITA–1 \(\%\) DESI–East samples: \(F_{X,0.5-2}\geqslant 4\times 10^{-14}\) erg s \({}^{-1}\) cm \({}^{-2}\) (bright), \(F_{X,0.5-2}=(1.5{-}4)\times 10^{-14}\) erg s \({}^{-1}\) cm \({}^{-2}\) (medium), and \(F_{X,0.5-2}=(0.6{-}1.5)\times 10^{-14}\) erg s \({}^{-1}\) cm \({}^{-2}\) (faint) with a total area of 91.4, 91.4, and 16.62 deg \({}^{2}\) , respectively. We propose a new method of postprocessing the probabilistic photo-z predictions based on a two-temperature correction of the probability density function \(PDF(z)\) . This approach allows the calibration of the probabilistic predictions and confidence intervals of the photometric redshifts for eROSITA X-ray sources obtained by SRGz to be improved significantly.

中文翻译：

---

SRGz：1 $${\%}$$ DESI 区域中 eROSITA 点 X 射线源的分类和光度红移校准*

摘要

我们考虑了东银河半球1 \(\%\) DESI 光谱巡天区域 (eROSITA–1 \(\%\) DESI–East两年 SRG/eROSITA 巡天的点 X 射线源数量）。正在考虑的数据结合了大调查区域（与中格式 Stripe82X 和 XMM-XXL X 射线调查相比）和 X 射线源光学对应物的光谱学的高完整性。我们比较了三个 X 射线通量限制 eROSITA–1 \(\%\)中物体的类别和红移的光度 (SRGz) 和光谱/天体测量 (DESI EDR、SDSS、HELP、GAIA) 测量结果DESI–East 样本：\(F_{X,0.5-2}\geqslant 4\times 10^{-14}\) erg s \({}^{-1}\) cm \({}^{-2 }\)（亮），\(F_{X,0.5-2}=(1.5{-}4)\times 10^{-14}\) erg s \({}^{-1}\) cm \ ({}^{-2}\)（中），且\(F_{X,0.5-2}=(0.6{-}1.5)\times 10^{-14}\) erg s \({}^ {-1}\) cm \({}^{-2}\)（微弱），总面积分别为 91.4、91.4 和 16.62 deg \({}^{2}\)。我们提出了一种基于概率密度函数\(PDF(z)\)的两个温度校正来后处理概率 photo-z 预测的新方法。这种方法可以显着改善 SRGz 获得的 eROSITA X 射线源光度红移的概率预测和置信区间的校准。