AI 中文总结
该研究利用模拟为 CASTOR 和 NGRST 呈现星系合成图像，考虑多种因素后用\texttt{FAST++}等拟合光度观测，测量恒星形成活动指标，能区分熄灭星系及机制，用机器学习提高分类准确性，还对调查中熄灭群体丰度做了预测。
AI 中文摘要
我们基于 IllustrisTNG 套件的模拟，利用恒星群体合成库\textsc{galaxev}，呈现了处于恒星形成熄灭阶段的星系的合成图像，这些图像是针对宇宙学光学和紫外研究高级巡天望远镜（CASTOR）和南希·格雷斯·罗曼太空望远镜（NGRST）的。我们考虑了尘埃和各种噪声源的影响以产生模拟观测，这些观测应反映真实观测。使用这些合成图像，我们用\texttt{FAST++}和灵活的恒星形成历史对分箱圆形环带中的光度观测进行拟合，并很好地恢复了空间分辨的恒星质量和恒星形成率。由此我们测量了星系熄灭背景下空间分辨恒星形成活动的各种指标（形态学指标）。我们发现能够将熄灭星系与质量匹配的正常恒星形成星系控制样本区分开来。此外，我们还能区分各种熄灭机制，其中与由内而外熄灭特征一致的星系可以与显示由外而内特征的星系分开。使用机器学习技术，这种分类的准确性是可靠的，并且可以估计不同熄灭星系群体在熄灭过程中的进展。我们对 CASTOR 和 NGRST 拟议调查中各种熄灭群体的丰度进行了预测，发现这些调查将能够对数千个熄灭星系进行分类，直至中等红移，考虑更高红移时能分类更多。
英文摘要
We present synthetic images of galaxies that are in the stages of star formation quenching for the Cosmological Advanced Survey Telescope for Optical and UV Research (CASTOR) and Nancy Grace Roman Space Telescope (NGRST), based on simulations coming from the IllustrisTNG suite, as processed using the stellar population synthesis library \textsc{galaxev}. We account for the effects of dust and various sources of noise to produce mock observations that should mirror real observations. Using these synthetic images, we fit photometric observations in binned circular annuli using \texttt{FAST++} and a flexible star formation history, and recover well the spatially resolved stellar mass and star formation rate. We thereby measure various indicators (morphological metrics) of spatially resolved star formation activity in the context of galaxy quenching. We find that we are able to distinguish quenching galaxies from a mass-matched control sample of normal star forming galaxies. We additionally find that we can distinguish various quenching mechanisms, where galaxies consistent with an inside-out quenching signature can be separated from galaxies that display an outside-in signature. Using machine learning techniques the accuracy of this classification is reliable, and the progress through the quenching episode can be estimated for the different populations of quenching galaxies. We make predictions for the abundance of the various quenching populations in proposed surveys for CASTOR and NGRST, and find that these surveys will enable the classifications of thousands of quenching galaxies out to intermediate redshifts, and more when considering higher redshifts.
Comments23 pages, 12 figures, 3 tables. Submitted to OJAp
