发表机构
Università di Napoli “Federico II”; Istituto Nazionale di Fisica Nucleare, Sezione di Napoli; University of Tübingen（那不勒斯费德里科二世大学; 意大利国家核物理研究所，那不勒斯分部; 蒂宾根大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过校准eikonal方法结合贝叶斯推断，比较度规特定与度规无关模型对旋转黑洞光环局部几何的约束，发现多参数分析更鲁棒但存在简并，且均与Kerr度规一致。
AI 中文摘要
黑洞的准正则模为通过致密双星并合铃宕阶段的引力波观测来检验广义相对论提供了核心途径。然而，推断黑洞性质或背景时空需要依赖于理论的输入。在本工作中，我们将一种校准的eikonal方法（该方法将准正则模频移与光环附近的局部度规性质联系起来）与潜在的旋转黑洞度规相连接。利用贝叶斯推断，我们比较了基于全局Konoplya-Rezzolla-Zhidenko参数化的度规特定模型与基于光环处度规函数的局部值和径向导数的度规无关方法。对于每个模型，我们进行了单参数和同时多参数分析，并量化了它们的先验如何映射到相同的局部量上。我们发现，度规特定系数上的均匀先验可能在这些局部度规量上诱发强非均匀且偏斜的先验。对于与广义相对论兼容的注入，两个模型都与Kerr度规保持一致，尽管它们诱导的先验不同，但在轨道频率和Lyapunov指数上产生了相似的后验。对于代表偏离广义相对论的注入，单参数分析可能无法恢复注入的局部偏差，并且在度规特定情况下，可能使全局度规重建产生偏差。两个模型的多参数分析产生了一致的约束，表明具有足够的鲁棒性。然而，它们引入了实质性的简并和边缘化参数的差异，展示了在考虑更灵活模型时的实际局限性。最后，我们将该框架应用于GW250114铃宕分析的近似后验信息，发现两个模型都与广义相对论中的Kerr度规一致。
英文摘要
Quasinormal modes of black holes provide a central avenue for confronting general relativity with gravitational-wave observations from the ringdown stage of compact binary coalescence. However, inferring black-hole properties or the background spacetime requires theory-dependent input. In this work, we connect a calibrated eikonal method, which relates quasinormal mode shifts to local metric properties near the light ring, with the underlying rotating black hole metric. Using Bayesian inference, we compare a metric-specific model based on the global Konoplya-Rezzolla-Zhidenko parametrization with a metric-agnostic approach based on local values and radial derivatives of the metric functions at the light ring. For each model, we perform both one-parameter and simultaneous multi-parameter analyses and quantify how their priors map onto the same local quantities. We find that uniform priors on the metric-specific coefficients can induce strongly nonuniform and skewed priors on these local metric quantities. For an injection compatible with general relativity, both models remain consistent with Kerr and yield similar posteriors for the orbital frequency and Lyapunov exponent despite their different induced priors. For injections representing deviations from general relativity, one-parameter analyses can fail to recover the injected local deviations and, in the metric-specific case, can bias the global metric reconstruction. The multi-parameter analyses of both models yield consistent constraints, suggesting sufficient robustness. However, they introduce substantial degeneracies and differences in marginalized parameters, demonstrating practical limitations when considering more flexible models. Finally, we apply the framework to approximate posterior information from the ringdown analysis of GW250114 and find that both models are consistent with the Kerr metric in general relativity.
Comments20 pages, 14 figures
