This is the authors' abstract. We don't add key points for this paper.
As LEO megaconstellations such as Starlink become critical infrastructure for millions of users, it becomes important to examine the effects of space weather phenomena such as geomagnetic or solar storms. The network-level impact of these events remains poorly characterised, especially across orbit types and at application scale. Using a three-pronged measurement strategy that combines real-time controlled experiments via LEOScope, post-hoc analysis of over 4 million M-Lab NDT7 speedtests spanning 240 countries, and Cloudflare AIM application quality scores, we study the effects on Starlink of all the fifteen strong, severe, or extreme geomagnetic storms that occurred from May 2024 to October 2025. We find that LEO networks suffer severe, latitude-bounded degradation: packet loss doubles, throughput drops by 20–40%, and connection drops nearly quadruple. LEO is disproportionately affected relative to MEO and GEO. Application scores for gaming, streaming, and VoIP degrade by one quality tier. BBR consistently outperforms CUBIC under storm conditions. We also find that satellite operators manoeuvre nearly half their LEO fleets during storms, a necessary but performance-costly response. These findings call for space-weather-aware satellite network operations and end-user transparency which may allow for user-side adaptation during geomagnetic storms.
No takes yet. Share an insight, caveat, or question.
Fadaei et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: