论文标题

部分可观测时空混沌系统的无模型预测

Polarized Emission From Four Supernova Remnants In The THOR Survey

论文作者

Shanahan, Russell, Stil, Jeroen M., Anderson, Loren, Beuther, Henrik, Goldsmith, Paul, Ott, Juergen, Rugel, Michael, Soler, Juan, Syed, Jonas

论文摘要

储层计算是预测湍流的有力工具,其简单的架构具有处理大型系统的计算效率。然而,其实现通常需要完整的状态向量测量和系统非线性知识。我们使用非线性投影函数将系统测量扩展到高维空间,然后将其输入到储层中以获得预测。我们展示了这种储层计算网络在时空混沌系统上的应用,该系统模拟了湍流的若干特征。我们表明,使用径向基函数作为非线性投影器,即使只有部分观测并且不知道控制方程,也能稳健地捕捉复杂的系统非线性。最后,我们表明,当测量稀疏、不完整且带有噪声,甚至控制方程变得不准确时,我们的网络仍然可以产生相当准确的预测,从而为实际湍流系统的无模型预测铺平了道路。

We present polarization and Faraday rotation for the supernova remnants (SNRs) G46.8-0.3, G43.3-0.2, G41.1-0.3, and G39.2-0.3 in L-band (1-2 GHz) radio continuum in The HI/OH/Recombination line (THOR) survey. We detect polarization from G46.8-0.3, G43.3-0.2 and G39.2-0.3 but find upper limits at the 1% level of Stokes I for G41.1-0.3. For G46.8-0.3 and G39.2-0.3 the fractional polarization varies on small scales from 1% to ~6%. G43.3-0.2 is less polarized with fractional polarization <~3%. We find upper limits at the 1% level for the brighter regions in each SNR with no evidence for associated enhanced Faraday depolarization. We observe significant variation in Faraday depth and fractional polarization on angular scales down to the resolution limit of 16". Approximately 6% of our polarization detections from G46.8-0.3 and G39.2-0.3 exhibit two-component Faraday rotation and 14% of polarization detections in G43.3-0.2 are multi-component. For G39.2-0.3 we find a bimodal Faraday depth distribution with a narrow peak and a broad peak for all polarization detections as well as for the subset with two-component Faraday rotation. We identify the narrow peak with the front side of the SNR and the broad peak with the back side. Similarly, we interpret the observed Faraday depth distribution of G46.8-0.3 as a superposition of the distributions from the front side and the back side. We interpret our results as evidence for a partially filled shell with small-scale magnetic field structure and internal Faraday rotation.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源