Mapping circumgalactic medium observations to theory using machine learning
Mapping circumgalactic medium observations to theory using machine learning
复制标题
使用机器学习将环绕银河介质的观测结果映射到理论
DOI：
10.1093/mnras/stad2266
复制
发表时间：
2023
影响因子：
4.8
通讯作者：
Appleby S
中科院分区：
文献类型：
--
作者：
Appleby S
我们提出了一个随机森林（RF）框架，用于从类星体吸收线观测数据预测环星系介质（CGM）的物理条件，该框架在模拟宇宙学模拟的Voigt剖面拟合合成吸收器样本上进行了训练。传统上，从CGM吸收体观测中提取物理条件涉及简化假设，如均匀单相云，但通过使用宇宙学模拟，我们绕过了这些假设，以更好地捕捉CGM观测值与潜在气体条件之间的复杂关系。我们在合成光谱上训练RF模型，并在一系列恒星形成速率、恒星质量和撞击参数中选择星系周围的金属线，以预测吸收剂的过密度、温度和金属丰度。这些模型重现了simbawell的真实值，所有离子的过度密度的标准化横向标准差为0.50-0.54指数，温度的标准化横向标准差为0.32-0.54指数，金属线（不是Hi）预测的金属丰度的标准化横向标准差为0.49-0.53指数。通过对特征重要性的检验，RF表明，过密度与吸收塔密度最相关，温度受线宽驱动，金属丰度对特定恒星形成速率最敏感。或者，通过一次移除一个可观察到的特征来检查特征的重要性，过度密度和金属丰度似乎更多地由冲击参数驱动。为了保证网络准确地跨越真实物理条件下的散射，我们引入了一种归一化流方法。经过训练的模型可以在网上找到。
We present a random forest (RF) framework for predicting circumgalactic medium (CGM) physical conditions from quasar absorption line observables, trained on a sample of Voigt profile-fit synthetic absorbers from thesimbacosmological simulation. Traditionally, extracting physical conditions from CGM absorber observations involves simplifying assumptions such as uniform single-phase clouds, but by using a cosmological simulation we bypass such assumptions to better capture the complex relationship between CGM observables and underlying gas conditions. We train RF models on synthetic spectra for Hiand selected metal lines around galaxies across a range of star formation rates, stellar masses, and impact parameters, to predict absorber overdensities, temperatures, and metallicities. The models reproduce the true values fromsimbawell, with normalized transverse standard deviations of 0.50–0.54 dex in overdensity, 0.32–0.54 dex in temperature, and 0.49–0.53 dex in metallicity predicted from metal lines (not Hi), across all ions. Examining the feature importance, the RF indicates that the overdensity is most informed by the absorber column density, the temperature is driven by the line width, and the metallicity is most sensitive to the specific star formation rate. Alternatively examining feature importance by removing one observable at a time, the overdensity and metallicity appear to be more driven by the impact parameter. We introduce a normalizing flow approach in order to ensure the scatter in the true physical conditions is accurately spanned by the network. The trained models are available online.