A Deep Dive into the Impact of Solar Storms on LEO Satellite Networks¶

Low Earth Orbit (LEO) satellite networks are an important part of the global communication infrastructure today. Despite ongoing efforts to improve their resilience, they remain vulnerable to component damage and deorbiting under harsh space weather conditions. Prior work identified a modest but noticeable impact on LEO satellite network performance during solar storms, typically manifesting as an immediate rise in packet loss and a sustained increase in round-trip time (RTT). However, these studies offer only coarse-grained insights and do not capture the nuanced spatial and temporal patterns of disruption across the LEO network.

In this paper, we conduct a deep dive into the impact of solar storms on LEO satellite communications. By localizing the impact of increased atmospheric drag at the level of individual satellites and orbits, we reveal significant heterogeneity in how different parts of the network are affected. We find that the degree of performance degradation varies significantly across geographic regions, depending on satellite positioning during the storm. Specifically, we find that (i) not all satellite orbits are equally vulnerable, (ii) within a given orbit, certain satellites experience disproportionate impact depending on their position relative to geomagnetic conditions, and (iii) autonomous maneuvering of satellites might be a cause of the sustained increase in RTT. Our findings uncover previously overlooked patterns of vulnerability in LEO satellite constellations and highlight the need for more adaptive, region-aware mitigation strategies to address space weather-induced network disruptions.

如今，低地球轨道 (LEO) 卫星网络已成为全球通信基础设施的重要组成部分。尽管业界不断努力提升其鲁棒性，但在恶劣的空间天气条件下，它们仍然面临组件损坏和脱轨的风险。先前的工作已发现太阳风暴对 LEO 卫星网络性能存在中等但显著的影响，通常表现为丢包率的瞬时上升和往返时间 (RTT) 的持续增加。然而，这些研究仅提供了粗粒度的见解，未能捕捉到 LEO 网络中干扰事件在空间和时间上的细微模式。

在本文中，我们对太阳风暴对 LEO 卫星通信的影响进行了深入研究。通过在单个卫星和轨道的层面上定位大气阻力增加所带来的影响，我们揭示了网络不同部分受影响的显著异构性。

我们发现，性能下降的程度因地理区域而异，具体取决于卫星在风暴期间的位置。

具体而言，我们发现：

并非所有卫星轨道都同样脆弱
在同一轨道内，部分卫星根据其相对于地磁条件的位置会受到不成比例的巨大影响
卫星的自主机动可能是导致 RTT 持续增加的原因

我们的研究结果揭示了 LEO 卫星星座中先前被忽视的脆弱性模式，并强调需要采取更具适应性、感知区域差异的缓解策略，以应对空间天气引发的网络中断。

Introduction¶

Low Earth Orbit (LEO) satellite constellations are rapidly transforming global connectivity. With their ability to provide low-latency, wide-area coverage, LEO networks have become foundational to applications ranging from consumer Internet access to disaster response and maritime communications. Major initiatives such as Starlink [28] and OneWeb [25] have already deployed thousands of satellites, with plans for tens of thousands more. As dependence on LEO-based communication grows, ensuring its robustness becomes increasingly critical.

A major threat to LEO satellite operations is space weather. Satellites are directly exposed to extreme space weather events, such as solar flares and Coronal Mass Ejections (CMEs) [17]. These events, driven by increased solar activity, can elevate atmospheric drag at orbital altitudes, stressing hardware systems and affecting satellite trajectories [6, 8–10, 14, 15, 19, 24]. While significant engineering efforts have gone into improving satellite resilience, operational data from recent storms reveal persistent vulnerabilities, including network performance degradation and, in extreme cases, satellite deorbiting [23].

Prior work has shown that solar storms can cause modest yet measurable degradation in communication performance, typically manifesting as increased packet loss and elevated Round-Trip Time (RTT) [27]. However, past work tends to view the LEO network in aggregate, offering only coarse-grained insights. They do not account for the complex spatial and temporal dynamics of how different satellites and regions are affected during a storm. This gap limits our ability to design targeted mitigation strategies and adaptive communication protocols.

In this paper, we present a fine-grained analysis of solar storm impacts on LEO satellite networks. Using real-world network measurement data from RIPE probes connected to the Starlink network (AS 14593) and publicly available satellite position information across four solar storms in 2024, we investigate how atmospheric drag from geomagnetic disturbances affects different parts of the LEO constellation, and in turn, network performance.

Our study reveals significant heterogeneity in the impact based on the time and location of the storm. (i) Geographic variation: There are variations in network performance degradation across regions in terms of the time of peak impact. (ii) Orbital disparities: Not all orbital planes experience the same level of disruption during a storm. The extent of impact depends on the orientation of the orbit with respect to the sun during storm impact. (iii) Location-based variations: We recognize three broad location-based categories for highly impacted satellites. We identify relatively large altitude changes for satellites at high latitudes, those over the South Atlantic Anomaly (SAA) region [7], and those facing the sun during the time of solar storm impact. Additionally, we also uncover patterns in highly affected satellites across consecutive storm days, showing the propagation of impact across neighboring satellites.

By uncovering spatial patterns of vulnerability, our work provides actionable insights to improve the robustness of LEO communication systems. We also discuss implications for adaptive routing and long-term constellation design in the face of increasing solar activity.

低地球轨道 (LEO) 卫星星座正在迅速改变全球的连接方式。凭借其提供低延迟、广域覆盖的能力，LEO 网络已成为从消费级互联网接入到灾难响应和海事通信等多种应用的基础。Starlink [28] 和 OneWeb [25] 等主要项目已经部署了数千颗卫星，并计划再部署数万颗。随着对 LEO 通信的依赖日益增长，确保其鲁棒性变得至关重要。

空间天气是 LEO 卫星运营面临的一个主要威胁。卫星直接暴露在极端空间天气事件中，如太阳耀斑和日冕物质抛射 (CMEs) [17]。这些由太阳活动增强驱动的事件会增加轨道高度的大气阻力，给硬件系统带来压力并影响卫星轨道 [6, 8–10, 14, 15, 19, 24]。尽管在提高卫星鲁棒性方面已投入大量工程努力，但近期风暴的运行数据显示，脆弱性依然存在，包括网络性能下降，以及在极端情况下卫星脱轨 [23]。

先前的工作表明，太阳风暴会导致通信性能出现中等但可测量的下降，通常表现为丢包率增加和往返时间 (RTT) 升高 [27]。然而，过去的工作倾向于将 LEO 网络作为一个整体来看待，仅提供粗粒度的见解，并未考虑风暴期间不同卫星和区域受影响的复杂空间和时间动态。这一研究空白限制了我们设计有针对性的缓解策略和自适应通信协议的能力。

在本文中，我们对太阳风暴对 LEO 卫星网络的影响进行了细粒度分析。我们利用连接到 Starlink 网络 (AS 14593) 的 RIPE 探测器的真实网络测量数据，以及 2024 年四次太阳风暴期间公开的卫星位置信息，研究了地磁扰动引起的大气阻力如何影响 LEO 星座的不同部分，进而影响网络性能。

我们的研究揭示了基于风暴时间和地点的显著影响异构性。

(i) 地理差异：不同区域在网络性能下降的峰值影响时间上存在差异

(ii) 轨道差异：并非所有轨道平面在风暴期间都经历相同程度的中断。影响程度取决于风暴冲击时轨道相对于太阳的朝向

(iii) 基于位置的差异：我们识别出受影响严重卫星的三大类基于位置的特征

我们发现，位于高纬度地区、南大西洋异常区 (SAA) [7] 上空以及在太阳风暴冲击时朝向太阳的卫星，其高度变化相对较大。此外，我们还揭示了在连续的风暴日中，受严重影响的卫星群体的模式，显示了影响在相邻卫星间的传播。

通过揭示脆弱性的空间模式，我们的工作为提高 LEO 通信系统的鲁棒性提供了可行的见解。我们还讨论了在太阳活动日益增加的背景下，这对自适应路由和长期星座设计的影响。

Background¶

We provide a brief overview of prior work on the analysis of satellite network performance during solar storms.

While several works [16, 18, 21, 22, 26] have evaluated LEO satellite performance under normal operational conditions, the analysis under solar storms is limited. Recent work [27] explores the impact of the May 2024 storm on the Starlink network, highlighting the immediate impact on loss and delayed but sustained impact on RTTs. Furthermore, CosmicDance [9] uses real-world measurements to investigate the impact of solar storms on satellite orbits. However, both works stop short of linking orbital dynamics during solar storms with real-time network performance degradation.

Atmospheric drag [1] plays a critical role in satellite behavior, particularly during periods of heightened solar activity. Increased solar radiation heats the Earth’s upper atmosphere, causing it to expand and increase in density at higher altitudes. This denser atmosphere results in greater drag on LEO satellites, which in turn leads to orbital decay, often observed as a loss of altitude. Satellites must frequently compensate for this decay through propulsion maneuvers to maintain operational altitudes and prevent premature re-entry.

我们简要回顾先前关于太阳风暴期间卫星网络性能分析的工作。

虽然已有多项工作 [16, 18, 21, 22, 26] 在正常运行条件下评估了 LEO 卫星的性能，但关于太阳风暴下的分析还很有限。最近的工作 [27] 探讨了 2024 年 5 月风暴对 Starlink 网络的影响，强调了其对丢包率的即时影响以及对 RTT 的延迟但持续的影响。此外，CosmicDance [9] 利用真实世界的测量数据研究了太阳风暴对卫星轨道的影响。然而，这两项工作都未能将太阳风暴期间的轨道动力学与实时的网络性能下降联系起来。

大气阻力 [1] 在卫星行为中扮演着关键角色，尤其是在太阳活动加剧期间。增强的太阳辐射会加热地球高层大气，导致其膨胀并在更高的高度密度增加。这种更密集的大气对 LEO 卫星产生更大的阻力，进而导致轨道衰减，通常表现为高度的损失。卫星必须通过推进器机动频繁地补偿这种衰减，以维持运行高度并防止过早再入大气层。

Discussion¶

In this section, we discuss the implications of our findings on network design and resilience of LEO constellations.

在本节中，我们讨论我们的发现对 LEO 星座网络设计和鲁棒性的启示。

Leveraging Predictability of Impact: Our findings demonstrate that satellites at high latitudes and those with direct solar exposure during storm peaks experience the most significant orbital decay, corroborating previous thermosphere ionosphere coupled simulations [12]. This predictability presents critical opportunities for enhancing network resilience through proactive management strategies.

Operators can implement preemptive network adaptations by integrating solar storm forecasts with our identified vulnerability patterns. Networks can dynamically reconfigure topologies, preposition backup communication paths through less-vulnerable satellites, and implement priority-based handover mechanisms that favor satellites likely to be in low-impact orbits. By predicting when and where satellites are likely to experience orbital degradation, networks can dynamically adapt routing protocols and create resilience zones by identifying satellite clusters likely to maintain stable performance during storms.

However, not all satellites traversing high-risk regions experience uniform altitude degradation. This variability is due to highly localized atmospheric density deviations and satellite-specific factors. This heterogeneity necessitates the development of probabilistic impact models that capture the stochastic nature of drag effects, enabling networks to make risk-aware routing decisions that balance performance optimization with reliability guarantees during space weather events.

利用影响的可预测性：我们的研究结果表明，高纬度地区和在风暴高峰期直接面向太阳的卫星经历了最显著的轨道衰减，这与先前的热层-电离层耦合模拟结果 [12] 相符。这种可预测性为通过主动管理策略增强网络鲁棒性提供了重要机会。

运营商可以通过将太阳风暴预报与我们识别出的脆弱性模式相结合，实施先发制人的网络调整。网络可以动态地重构拓扑结构，通过受影响较小的卫星预先部署备用通信路径，并实施基于优先级的切换机制，优先选择可能处于低影响轨道的卫星。通过预测卫星可能在何时何地经历轨道衰减，网络可以动态调整路由协议，并通过识别可能在风暴期间保持稳定性能的卫星集群来创建“弹性区域”。

然而，并非所有穿越高风险区域的卫星都经历相同的高度下降。这种差异是由于高度局部化的大气密度偏差和卫星特定因素造成的。这种异构性要求我们开发能够捕捉阻力效应随机性的概率影响模型，使网络能够在空间天气事件期间做出兼顾性能优化和可靠性保障的风险感知路由决策。

The Propagation of Impact across Satellites: Our analysis reveals that different sets of satellites experience the most severe impacts across consecutive storm days, with only 30 out of 236 highly impacted satellites on May 11 th remaining in the high-impact category on May 12 th . While atmospheric density pockets could theoretically affect multiple satellites across days, the observed pattern of neighboring satellites being sequentially impacted suggests a different cause. We attribute this clustered propagation primarily to the self-driving nature of LEO constellations. Starlink employs a proprietary onboard autonomous control system to manage continuous orbit maintenance, collision avoidance, and inter-shell maneuvers. However, excessive or unnecessary maneuvers can lead to network topology instability and performance degradation [20]. Analysis of TLE data reveals that when satellites experience altitude loss due to increased atmospheric drag, Starlink responds by temporarily raising the affected satellites above their nominal altitude. These satellites typically return to their original altitude within 1–2 days. This corrective action triggers a cascading effect, with orbital adjustments propagating across neighboring satellites in both spatial and temporal dimensions. Full stabilization of the orbit often takes 3–4 days. These dynamic adjustments can disrupt satellite links and routing paths, contributing to performance issues such as a sustained increase in round-trip time (RTT).

This finding raises fundamental questions about the suitability of today’s autonomous constellation management during extreme space weather events. The self-driving algorithms, optimized for normal operations, may inadvertently amplify storm impacts by triggering chains of orbital adjustments. Network operators should investigate implementing storm-aware autonomous control modes that temporarily modify or disable certain self-driving behaviors during severe geomagnetic disturbances and introduce collective decision-making algorithms that consider space weather conditions when planning constellation-wide maneuvers.

影响在卫星间的传播：我们的分析显示，在连续的风暴日中，受影响最严重的卫星群体是不同的，5 月 11 日的 236 颗高影响卫星中，只有 30 颗在 5 月 12 日仍属于高影响类别。虽然理论上大气密度团块可能连续几天影响多颗卫星，但观测到的相邻卫星相继受影响的模式指向了另一个原因。我们主要将这种集群式传播归因于 LEO 星座的“自驾驶”特性。Starlink 采用专有的星上自主控制系统来管理持续的轨道维持、碰撞规避和壳层间机动。然而，过度或不必要的机动可能导致网络拓扑不稳定和性能下降 [20]。对 TLE 数据的分析显示，当卫星因大气阻力增加而经历高度损失时，Starlink 的应对措施是暂时将受影响的卫星提升到其标称高度之上。这些卫星通常在 1-2 天内返回原始高度。这种纠正措施会引发级联效应，轨道调整在空间和时间维度上会传播到相邻卫星。轨道的完全稳定通常需要 3-4 天。这些动态调整会干扰卫星链路和路由路径，从而导致性能问题，例如往返时间 (RTT) 的持续增加。

这一发现对当今自主星座管理系统在极端空间天气事件下的适用性提出了根本性问题。为正常运行而优化的自驾驶算法可能会因触发一系列轨道调整而无意中放大风暴的影响。网络运营商应研究实施“风暴感知型自主控制模式”，在严重地磁扰动期间暂时修改或禁用某些自驾驶行为，并引入在规划星座范围内的机动时考虑空间天气条件的集体决策算法。

A Real-time Monitoring Framework: A monitoring framework that compares predicted solar storm impact models against observed network performance metrics can enable the iterative refinement of probabilistic impact models in real time, in turn facilitating network adaptations. For example, such a monitoring system could provide early warning capabilities by detecting initial impacts in eastern regions before they propagate westward. This real-time validation loop would improve the accuracy of the 3-day forecast window typically available for solar storm predictions, allowing operators to make more informed decisions about preemptive network reconfigurations and resource allocation.

实时监控框架：一个能够将预测的太阳风暴影响模型与观测到的网络性能指标进行比较的监控框架，可以实现对概率影响模型的实时迭代优化，从而促进网络调整。例如，这样的监控系统可以在影响向西传播之前，通过检测东部地区的初始影响来提供预警能力。这种实时验证回路将提高太阳风暴预报通常提供的 3 天预测窗口的准确性，使运营商能够就先发制人的网络重构和资源分配做出更明智的决策。

Limitations: Our analysis is constrained by the sparsity of available data sources. The satellite TLE data, with an average gap of 10 hours between reported positions, makes it difficult to determine the precise timing of satellite decay. Similarly, latency analysis is limited by the coarse resolution of ping measurements, taken every 4 minutes, which fail to capture short-term variations in loss and latency. Additionally, the sparse distribution of measurement probes at lower latitudes results in reduced data coverage in those regions.

The opaque nature of the Starlink system further constrains our analysis. Although RIPE Atlas measurements provide the geographic coordinates of probes, they do not reveal which satellite is serving a given probe at any point in time. Additionally, the lack of traceroute support within the Starlink network prevents us from identifying the satellite hops involved along a given path.

The observed impact of solar storms on satellite performance is also highly non-uniform. While a higher concentration of affected satellites appears at higher latitudes, not all satellites in these regions are impacted. Even in the most affected regions, no more than 20% of satellites experienced significant altitude changes. This uneven impact is likely due to local variations in atmospheric conditions along individual satellite paths. These findings underscore the need for a more detailed analysis to gain a deeper understanding of the fine-grained dynamics of satellite behavior under solar storm conditions.

局限性：我们的分析受限于可用数据源的稀疏性。卫星 TLE 数据的报告位置之间平均有 10 小时的间隔，这使得我们难以确定卫星衰减的精确时间。同样，延迟分析也受限于每 4 分钟一次的 ping 测量，其粗糙的分辨率无法捕捉到丢包和延迟的短期变化。此外，测量探测器在低纬度地区的稀疏分布导致了这些区域的数据覆盖率降低。

Starlink 系统的不透明性进一步限制了我们的分析。尽管 RIPE Atlas 测量提供了探测器的地理坐标，但它们并未揭示在任何时间点是哪颗卫星在为特定探测器提供服务。此外，Starlink 网络内部缺乏 traceroute 支持，使我们无法识别给定路径上的卫星跳。

观测到的太阳风暴对卫星性能的影响也高度不均匀。虽然受影响的卫星更多地集中在高纬度地区，但并非该区域的所有卫星都受到影响。即使在受影响最严重的区域，经历显著高度变化的卫星也不超过 20%。这种不均匀的影响可能是由于单个卫星路径上局部大气条件的变化所致。这些发现强调，需要进行更详细的分析，以更深入地了解太阳风暴条件下卫星行为的细粒度动态。

Conclusion¶

In this paper, we present a fine-grained analysis of the impact of solar storms on LEO networks by correlating real-world network performance data with satellite orbital dynamics across multiple storm events. Our study revealed significant spatial, temporal, and orbital heterogeneity in how solar activity impacts LEO network performance, highlighting vulnerable satellite regions, including high latitudes, sun-facing orbits, and the South Atlantic Anomaly. Future work should explore the design of adaptive network solutions as well as cross-layer control loops that integrate atmospheric drag forecasts, real-time RTT feedback, and autonomous satellite maneuver planning to build next-generation resilient satellite networks. While this study focuses on Starlink—the largest LEO constellation currently in operation—our methodology can be extended to analyze other systems such as OneWeb [25] and Kuiper [13].

在本文中，我们通过将真实世界的网络性能数据与多个风暴事件中的卫星轨道动力学相关联，对太阳风暴对 LEO 网络的影响进行了细粒度分析。我们的研究揭示了太阳活动影响 LEO 网络性能在空间、时间和轨道上的显著异构性，并指出了脆弱的卫星区域，包括高纬度地区、面向太阳的轨道和南大西洋异常区。未来的工作应探索设计自适应网络解决方案以及跨层控制回路，该回路应整合大气阻力预报、实时 RTT 反馈和自主卫星机动规划，以构建下一代高鲁棒性的卫星网络。虽然本研究聚焦于当前运营中最大的 LEO 星座 Starlink，但我们的方法可以扩展到分析其他系统，如 OneWeb [25] 和 Kuiper [13]。