Abstract 40Ar/39Ar geochronology is a powerful technique for dating geological events and processes on timescales from hundreds to billions of years. Many 40Ar/39Ar datasets are collected from analysis of single mineral phases or phenocryst-free groundmass that cooled rapidly following a volcanic eruption, which can allow for straightforward interpretation of 40Ar/39Ar age spectra. However, 40Ar/39Ar age spectra from mixtures of multiple minerals and/or multiple age components are often complex. In such situations, interpretations commonly used for single mineral phases are inappropriate and will result in geologically spurious conclusions. Here, we present a Bayesian method for the analysis and interpretation of 40Ar/39Ar step-heating spectra that result from mixing of multiple components, where a component is defined by both its age and mineral composition. We test the efficacy of this Bayesian approach using a suite of case studies. Two of these case studies utilize 40Ar/39Ar data from laboratory-prepared mixtures, which we use to explore how the composition, age, and number of components in a mixture, as well as our prior knowledge of these parameters, influence the model results. We also present an application-based case study in which we use plausible compositions and ages from a past Mars landing site to generate a synthetic 40Ar/39Ar dataset, which we then deconvolve using our Bayesian approach. We discuss modifications to our method that could improve the model's precision and outline geologic applications of our Bayesian approach in terrestrial and extraterrestrial settings that would permit the extraction of a greater amount of temporal information.

中文翻译：

---

从矿物混合物中解卷积 40Ar/39Ar 数据的贝叶斯方法

摘要 40Ar/39Ar年代学是一种强大的技术，可以对数百至数十亿年的时间尺度上的地质事件和过程进行测年。许多 40Ar/39Ar 数据集是通过对火山爆发后迅速冷却的单一矿物相或无斑晶地块的分析收集的，这可以直接解释 40Ar/39Ar 年龄谱。然而，来自多种矿物和/或多种年龄成分的混合物的 40Ar/39Ar 年龄谱通常很复杂。在这种情况下，通常用于单一矿物相的解释是不合适的，并且会导致地质学上的虚假结论。在这里，我们提出了一种贝叶斯方法，用于分析和解释由多种成分混合产生的 40Ar/39Ar 步进加热光谱，其中一个成分由其年龄和矿物成分定义。我们使用一组案例研究来测试这种贝叶斯方法的有效性。其中两个案例研究利用来自实验室制备的混合物的 40Ar/39Ar 数据，我们用这些数据来探索混合物中的成分、年龄和组分数量，以及我们对这些参数的先验知识如何影响模型结果。我们还提供了一个基于应用的案例研究，其中我们使用过去火星着陆点的合理成分和年龄来生成合成的 40Ar/39Ar 数据集，然后我们使用贝叶斯方法对其进行去卷积。我们讨论了可以改进模型的方法的修改