arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.08207astro-ph.HE
收藏

Katu:一个快速开放源码的全轻强子动力学代码,适用于blazar的贝叶斯推断建模

Katu: a fast open-source full lepto-hadronic kinetic code suitable for Bayesian inference modelling of blazars

Bruno Jiménez Fernández, Hendrik van Eerten

摘要:

在blazar等辐射等离子体流的模拟中,正确地演化一系列耦合不同粒子种类的动力学方程是一个计算上具有挑战性的问题。在本文中,我们介绍的代码Katu,能够同时进化的大量粒子种类,包括光子,轻子和强子及其相互作用。对于每个粒子,动力学方程的数值表达式都经过了优化,并且代码编写得很容易扩展和修改。此外,我们将Katu与两个统计软件emcee和pymultinest联系在一起,这使得一个套件可以应用于耀变体并从其电磁(和中微子,如果适用的话)通量中提取相关的统计信息。作为一个例子,我们应用Katu Mrk 421和比较轻子和轻强子模型。在这第一次直接贝叶斯比较这些模型的完整动力学模拟,我们发现大量的证据(贝叶斯因子$\sim7$)纯粹从电磁波谱的细强子模型。该代码可以在LGPL许可下在https://github.com/hveerten/katu上在线获得。

英文摘要:

In the modelling of blazars and other radiating plasma flows, correctly evolving in time a series of kinetic equation coupled for different particle species is a computationally challenging problem. In this paper we introduce the code Katu, capable of simultaneously evolving a large number of particle species including photons, leptons and hadrons and their interactions. For each particle, the numerical expressions for the kinetic equations are optimized for efficiency and the code is written to easily allow for extensions and modifications. Additionally, we tie Katu to two pieces of statistical software, emcee and pymultinest, which makes a suite ready to apply to blazars and extract relevant statistical information from their electromagnetic (and neutrino, if applicable) flux. As an example, we apply Katu to Mrk 421 and compare a leptonic and a leptohadronic model. In this first ever direct Bayesian comparison between full kinetic simulations of these models, we find substantial evidence (Bayes factor $\sim7$) for the leptohadronic model purely from the electromagnetic spectrum. The code is available online at https://github.com/hveerten/katu under the LGPL license.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑