人工智能帮助预测太阳耀斑

AI to aid prediction of solar flares

AI to aid prediction of solar flares (nature.com)

Using a computer-aided technique, astrophysicists have identified unique patterns in the Sun’s active regions, magnetic regions known to generate strong magnetic fields1. Such magnetic regions occasionally produce solar flares that emit harmful radiation, including significant X-ray radiation.

Deciphering the patterns in the magnetic regions will improve the understanding and prediction of flares, which pose threats to astronauts and satellite-based telecommunications on Earth.

Observational studies had provided clues about the physical processes that lead to flares. These studies, however, lacked data on definite flare precursors needed to reliably predict flares.

Scientists from the Tata Institute of Fundamental Research (TIFR) in Mumbai, India, New York University Abu Dhabi in United Arab Emirates and Stanford University in the United States trained a machine to distinguish flare-producing magnetic regions from the non-flaring ones by using magnetic region properties.

The machine successfully classified 75 per cent of the magnetic field observations that are separated in time from flares by at least 72 hours.

The researchers, led by Duttaraj B. Dhuri, detected features that depend on electric currents and magnetic forces in the magnetic regions. They also identified a systematic increase in electric currents and magnetic forces in these regions days prior to flares.

Such a backdrop reveals that the magnetic regions remain in a flare-ready state for several days before and after flares. Flare precursors identified in this research could be used to design a robust system to forecast an imminent flare-producing solar storm, allowing us to take evasive action, says Dhuri.

利用计算机辅助技术，天体物理学家在太阳的活动区（已知能产生强磁场的磁区）中发现了独特的模式。这些磁区偶尔会产生太阳耀斑，释放出有害的辐射，包括大量的X射线辐射。

解读这些磁区中的模式将有助于更好地理解和预测耀斑，因为耀斑对宇航员和地球上的卫星通信构成了威胁。

观测研究已经提供了关于导致耀斑产生的物理过程的线索。然而，这些研究缺乏可靠预测耀斑所需的明确耀斑前兆的数据。

来自印度孟买的塔塔基础科学研究院（TIFR）、阿拉伯联合酋长国纽约大学阿布扎比分校和美国斯坦福大学的科学家训练了一台机器，利用磁区的特性来区分产生耀斑的磁区和不产生耀斑的磁区。

该机器成功地分类了至少与耀斑相隔72小时的75%的磁场观测数据。

由杜塔拉吉·B·杜里领导的研究人员检测到了磁区中依赖于电流和磁力的特征。他们还发现，在耀斑发生前几天，这些区域的电流和磁力出现了系统性增加。

这一背景表明，在耀斑发生前后的几天里，磁区都处于耀斑易发状态。杜里表示，本研究中识别的耀斑前兆可用于设计一个强大的系统来预测即将发生的产生耀斑的太阳风暴，从而让我们能够采取规避行动。

References

1. Dhuri, D. B. et al. Machine learning reveals systematic accumulation of electric current in lead-up to solar flares. Proc. Natl. Acad. Sci. Unit. States. Am. (2019) doi: 10.1073/pnas.1820244116

Machine learning reveals systematic accumulation of electric current in lead-up to solar flares | PNAS

Machine learning reveals systematic accumulation of electric current in lead-up to solar flares

Significance

Reliable flare forecasting is essential for improving preparedness for severe space weather consequences. Flares also serve as probes of solar magnetic processes and the emergence of flux at the solar surface. Training machine-learning (ML) algorithms using magnetic-field observations for improving flare forecasting has been extensively studied in prior literature. Instead, here we use ML to understand the underlying mechanisms governing flares. We train ML algorithms to classify flaring and nonflaring active regions (ARs) with high fidelity and report statistical trends for AR evolution days before and after M- and X-class flares. These trends are interpreted in terms of existing models of subsurface magnetic field and flux emergence. Our results also provide hypotheses for achieving reliable flare forecasting.

意义
可靠的耀斑预测对于提高严重空间天气后果的应对能力至关重要。耀斑也是探测太阳磁过程和太阳表面磁通浮现的探针。以往文献已广泛研究利用磁场观测训练机器学习（ML）算法以改进耀斑预测。然而，本文我们使用机器学习来理解控制耀斑产生的潜在机制。我们训练机器学习算法以高保真度分类耀斑活跃区和非耀斑活跃区，并报告M级和X级耀斑前后数天内活跃区演变的统计趋势。这些趋势根据现有的地下磁场和磁通浮现模型进行了解释。我们的研究结果还为实现可靠的耀斑预测提供了假设。

Abstract

Solar flares—bursts of high-energy radiation responsible for severe space weather effects—are a consequence of the occasional destabilization of magnetic fields rooted in active regions (ARs). The complexity of AR evolution is a barrier to a comprehensive understanding of flaring processes and accurate prediction. Although machine learning (ML) has been used to improve flare predictions, the potential for revealing precursors and associated physics has been underexploited. Here, we train ML algorithms to classify between vector–magnetic-field observations from flaring ARs, producing at least one M-/X-class flare, and nonflaring ARs. Analysis of magnetic-field observations accurately classified by the machine presents statistical evidence for (i) ARs persisting in flare-productive states—characterized by AR area—for days, before and after M- and X-class flare events; (ii) systematic preflare buildup of free energy in the form of electric currents, suggesting that the associated subsurface magnetic field is twisted; and (iii) intensification of Maxwell stresses in the corona above newly emerging ARs, days before first flares. These results provide insights into flare physics and improving flare forecasting.

意义
可靠的耀斑预测对于提高严重空间天气后果的应对能力至关重要。耀斑也是探测太阳磁过程和太阳表面磁通浮现的探针。以往文献已广泛研究利用磁场观测训练机器学习（ML）算法以改进耀斑预测。然而，本文我们使用机器学习来理解控制耀斑产生的潜在机制。我们训练机器学习算法以高保真度分类耀斑活跃区和非耀斑活跃区，并报告M级和X级耀斑前后数天内活跃区演变的统计趋势。这些趋势根据现有的地下磁场和磁通浮现模型进行了解释。我们的研究结果还为实现可靠的耀斑预测提供了假设。