Can solar eruptions be forecast using a novel combination of observations and machine learning techniques?

太阳的特征是由其动态和不断变化的磁场决定的，特别是在被称为活动区的强磁场区域。当活跃区域还很年轻的时候，它们是太阳系中最猛烈和最有活力的事件的来源-日冕物质抛射和太阳耀斑。当活跃区域死亡时，它们将磁场分散到整个太阳，这个磁场的残余物成为下一个太阳周期的输入。但是这些垂死的活跃区域继续产生日冕物质抛射。这些喷射是非常令人感兴趣的，因为它们可以驱动地球上的主要空间天气影响。例如，它们在近地空间环境中产生的变化最终可能导致配电、通信和导航系统中断。因此，了解这些喷射的原因和时间对于理解太阳如何运行以及开发准确的空间天气预报能力至关重要。目前，虽然我们对日冕物质抛射的理解已经取得了许多进展，但还没有可靠的方法来预测它们的发生。该项目将使用最先进的机器学习图像分析技术，并利用这些技术建立日冕物质抛射的预测模型。

The Sun's character is determined by its dynamic and evolving magnetic field, particularly in regions of intense magnetism known as active regions. When active regions are young, they are the source of the most violent and energetic events in the Solar System - coronal mass ejections and solar flares. When active regions die, they disperse their magnetic field across the Sun and remnants of this field become the input for the next solar cycle. But these dying active regions continue to produce coronal mass ejections. These ejections are of intense interest as they can drive major space weather impacts at Earth. For example, the changes they produce in the near-Earth space environment can ultimately lead to disruptions to electricity distribution, communications, and navigation systems. Knowing why and when these ejections will occur is therefore centrally important to understanding how the Sun operates but also for developing an ability to make accurate space weather forecasts. Currently, although many advances have been made in our understanding of coronal mass ejections, there is no reliable way to predict their occurrence in advance. This project will use state-of-the-art machine learning image analysis techniques and use these to build a predictive model for coronal mass ejections.