银河系的形成与演化是天体物理学中的核心问题。研究银河系演化的关键途径之一是分析其中不同星族的化学组成、动力学性质以及这些性质随时间的演化趋势，其中恒星年龄至关重要。通过恒星观测特征与恒星演化模型的对比，可以测定大样本恒星的年龄。恒星演化模型的重要输入参数之一是恒星的化学组成。传统的 增丰模型仅考虑了Mg、Si、Ca、Ti 四种 元素的加权平均丰度，而在恒星大气中占比更重的氧元素的观测丰度却没有被单独考虑。很多高分辨率光谱数据已经显示出氧元素丰度与其他 元素丰度之间存在显著差异，氧元素相比于其他 元素存在着明显的增丰和减丰现象。氧元素是质量丰度最大的金属元素，其丰度变化会对恒星的不透明度产生显著影响，进而影响恒星年龄的估算。为了获得更可靠的恒星年龄，需要单独考虑氧丰度变化来构建恒星模型，即构建氧增丰模型。本研究基于LAMOST 和GALAH 的光谱数据中 丰度和氧丰度的分布特征，构建了大规模的氧增丰模型数据库，并测定了矮星、主序拐点星和亚巨星样本的年龄，研究了年龄与化学丰度之间的关系。基于精确的恒星年龄，本文进一步探讨了银盘的化学演化过程以及银河系与矮星系的相互作用。本文的主要研究结果如下：

氧增丰模型被应用于67503 颗LAMOST DR5 矮星和4006 颗GALAH DR3矮星的年龄测定中。对于氧丰度大于 丰度的富氧星，基于氧增丰模型测定的年龄相比基于 增丰模型的年龄更小。而贫氧星则相反，基于氧增丰模型测定的贫氧星年龄更大。两种模型的年龄差异取决于恒星的[Fe/H] 和[O/] 值。当[O/]每变化0.2 dex 时，相对富金属的恒星（−0.2 < [Fe/H] < 0.2）年龄变化约10%，而相对贫金属的恒星（−1 < [Fe/H] < −0.2）年龄变化约15%。这项工作首次将氧增丰模型应用于矮星大样本的年龄测定中，定量地研究了氧丰度变化对矮星年龄测定造成的影响，揭示了清晰的年龄偏差随[Fe/H] 和[O/] 的变化趋势。

在第一项工作基础上，氧增丰模型被进一步应用于41034 颗GALAH DR3 的亚巨星和主序拐点星的年龄测定中，获得了拥有高精度年龄的样本，样本年龄误差的中值为9.4%。这项工作对比了 增丰模型和氧增丰模型给出的年龄测定结果，发现当[O/] = −0.2 时，两种模型造成的贫氧星的平均年龄偏差为3.7%，而富氧星在[O/] = 0.2 时的平均年龄偏差为−5.3%，在[O/] = 0.4 时为−11.0%。这种年龄差异显著影响了厚盘和银晕恒星的年龄分布，导致[Fe/H]-年龄关系从8 Gyr到14 Gyr 表现出更陡峭的下降趋势，表明这些星族的形成时间尺度较短，化学增丰过程较快。该工作证实了薄盘恒星的归一化年龄-金属丰度分布(∣[Fe/H]) 存在V 形结构。此外，样本中的晕星可以分为两个序列，一个是富金属序列（Splash星），另一个是贫金属序列（吸积晕星）。其中大多数Splash 星的年龄超过9 Gyr，而吸积晕星的年龄超过10 Gyr。这项工作基于氧增丰模型重新限制了银河系中不同组分的年龄，清晰地刻画了薄盘、厚盘和银晕的年龄-金属丰度关系，展现了氧增丰模型在银河系考古学中的重要作用。

在第二项工作的基础上扩展了恒星样本，基于GALAH DR3 的43590 颗主序拐点星和亚巨星，探讨了银盘的金属丰度和氧丰度随时间的演化，并研究了银盘的径向化学丰度梯度的变化。这项工作同时考虑了径向迁移和矮星系吸积对银盘化学演化的影响，并将其与数值模拟结果进行了对比。该工作揭示了银盘中年龄-[Fe/H] 关系中的V 形结构是依赖于银河系中的位置的。当相对于银盘面的垂向距离zmax < 0.4 kpc 时，这种结构变得不连续，被4 Gyr 前至2 Gyr 前的金属丰度（[Fe/H]）下降和氧丰度（[O/Fe]）的显著增加过程所打断。通过计算样本恒星的出生半径，这项工作发现这些年轻的富氧恒星主要来自外盘，具有较大的出生半径，并通过径向迁移逐渐移动到太阳邻域。这些结果与前人的数值模拟结果一致，其模拟假设了一个质量与人马座矮星系相似的矮星系被银河系吸积的过程（矮星系与银河系的质量比约为0.001）。此外，这种矮星系吸积事件会导致径向金属丰度（[Fe/H]）和氧丰度（[O/Fe]）分布在2-4 Gyr 前的时期内产生明显变化。在这个时期内，随着年龄减小，径向[O/Fe] 梯度周围的[O/Fe] 弥散显著增加。这些发现表明，人马座矮星系可能会在外盘触发恒星形成爆发，产生年轻富氧星，并通过这些恒星的径向迁移重塑了银盘的化学丰度。这项工作基于氧增丰模型发现了薄盘中的年轻富氧星族，揭示了这种星族的外盘起源，并使用数值模拟成功解释了这个观测现象。

The formation and evolution of the Milky Way constitute a critical question in astrophysics. One of the key approaches to studying the evolution of the Milky Way is the analysis of the chemical compositions and dynamical properties of its various stellar populations, as well as the evolutionary trends of these properties over time, with stellar age being of paramount importance. By comparing observational properties of stars with stellar evolution models, the ages of a large sample of stars can be determined. A critical input for stellar evolution models is the chemical composition of stars. Traditional -enhanced stellar models only consider the error-weighted mean of four -elements abundances ([Mg/Fe], [Si/Fe], [Ca/Fe], and [Ti/Fe]), while the observed abundance of oxygen, which constitutes a larger proportion in the stellar atmosphere, is not separately considered. Many high-resolution spectroscopic data have revealed significant differences between the abundance of oxygen and that of other -elements, indicating notable enrichment or deficiency of oxygen abundance relative to other -elements abundance. Oxygen, being the most abundant metal element by mass, its abundance significantly affects the opacity of stars, thereby influencing the estimation of stellar ages. To obtain more reliable stellar ages, it is necessary to independently consider the variation of oxygen abundance to construct stellar models, namely oxygen-enhanced models. In this study, based on the distribution of abundance and oxygen abundance in the spectroscopic data from LAMOST and GALAH, we constructed a large-scale oxygen-enhanced model database and determined the ages of dwarf stars, main-sequence turnoff stars, and subgiants, investigating the relations between age and chemical abundance. Based on precise stellar ages, this paper further explores the chemical evolution of the Galactic disk and the interactions between the Milky Way and dwarf galaxies. The main findings of this paper are as follows:

The oxygen-enhanced models have been applied to age determinations of 67,503 LAMOST DR5 dwarf stars and 4,006 GALAH DR3 dwarf stars. For oxygen-rich stars with oxygen abundance greater than abundance, ages determined based on the oxygen-enhanced models are younger compared to those based on the -enhanced models. Conversely, for oxygen-poor stars, the ages determined based on the oxygen-enhanced models are older. The age difference between the two models depends on the [Fe/H] and [O/] values of the stars. Generally, varying 0.2 dex in [O/] will alter the age estimates of metal-rich (−0.2 < [Fe/H] < 0.2) stars by ∼10%, and relatively metal-poor (−1 < [Fe/H] < −0.2) stars by ∼15%. The maximum age difference between the two stellar models for oxygen-poor stars reaches 27%, and for oxygen-rich stars, it is −42%. This work represents the first application of the oxygen-enhanced models to age determination in a large sample of dwarf stars, quantitatively studying the impact of oxygen abundance variations on dwarf star age determination and revealing clear relations of age biases with [Fe/H] and [O/].

Beyond the first study, the oxygen-enhanced models were further employed for age determination of 41,034 subgiants and main-sequence turnoff stars in GALAH DR3, yielding a sample with high-precision ages, with a median age uncertainty of ∼9.4%. A comparative analysis of age determinations obtained from the -enhanced models and the oxygen-enhanced models revealed that at [O/] = −0.2, the average age difference between the two models for oxygen-poor stars is 3.7%, while for oxygen-rich stars at [O/] = 0.2, the average age difference is −5.3%, and at [O/] = 0.4, it is −11.0%. This age difference significantly impacts the age distributions of thick disc and halo stars, leading to a steeper downward trend in the [Fe/H]-age plane from 8 Gyr to 14 Gyr, indicating a shorter formation time-scale and a faster chemical enrichment history for these populations. This work confirms the V-shape of the normalized age-metallicity distribution (∣[Fe/H]) of thin disc stars. Furthermore, halo stars in the sample can be classified into two sequences, one being the metal-rich sequence (Splash stars) and the other being the metal-poor sequence (accreted halo stars). Most Splash stars age older than 9 Gyr, while accreted halo stars have ages older than 10 Gyr. This study, based on the oxygen-enhanced models, provides new age distributions for different components of the Milky Way, characterizing the age-metallicity relations of the thin disk, thick disk, and halo, thereby showcasing the crucial role of the oxygen-enhanced models in Galactic archaeology.

We extended the star sample from the second study, utilizing 43,590 main-sequence turnoff stars and subgiants from GALAH DR3, to investigate the evolution of metallicity and oxygen abundance in the Galactic disk over time, and to study the variations in the radial chemical abundance gradients of the Galactic disk. This work considered the effects of radial migration and accretion of dwarf galaxies on the chemical evolution of the Galactic disk and compared them with simulations. This study revealed that the V-shape structure of age-[Fe/H] relation depends on the location in the Galactic disc. This structure becomes discontinuous at z_max < 0.4 kpc, interrupted by a decrease in metallicity ([Fe/H]) and a significant increase in oxygen abundance from 4 Gyr to 2 Gyr ago. By calculating the birth radii of sample stars, this study found that these young O-rich stars have larger birth radii, originating predominantly from the outer disc, and subsequently migrating towards the Solar neighborhood through radial migration. These results is consistent with the simulation from previous work, which assumed a late satellite infall with a mass radio similar as that of the Sagittarius dwarf to the MW (∼0.001). Furthermore, this satellite infall could give rise to a distinct radial [Fe/H]/[O/Fe] profile compared to the earlier stage. The dispersion around the radial [O/Fe] gradient exhibits a remarkable increase as age decreases within the same epoch (2-4 Gyr). These findings indicate that Sagittarius dwarf galaxy can trigger star formation burst in the outer disc, contributing to the formation of young O-rich stars, and reshaping the chemical abundance of the Milky Way disc by radial migration. This work, based on oxygen-enhanced models, identifies a population of young oxygen-rich stars in the thin disk, revealing their outer disk origins and successfully explaining this observational phenomenon using numerical simulations.