Cosmic Ray Inter-Station Fluctuations as Precursors of Geomagnetic Storms: Causal Baseline Analysis and Forecasting Limitations
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.
Vera: We're looking at the paper: "Cosmic Ray Inter-Station Fluctuations as Precursors of Geomagnetic Storms".
Jocelyn: Here's the gist of the paper. Cosmic Ray Inter-Station Correlation Variations as Precursors of Geomagnetic Storms: A Statistical Study and Multi-Parameter Early Warning Framework analyzes the relationship between cosmic ray characteristics and geomagnetic storm intensity to develop a complementary predictive capability beyond existing L1 solar wind monitoring Geomagnetic storms (GSs) are significant manifestations of space weather arising from interactions between solar wind disturbances and Earth’s magnetosphere These events pose substantial risks to technological infrastructure, including satellite operations, power grid stability, and navigation systems GSs are driven by two primary solar wind structures: coronal mass ejections (CMEs) and high-speed streams (HSSs) emanating from coronal holes CMEs are transient, explosive eruptions that carry massive magnetized plasma structures through interplanetary space In contrast, HSSs are quasi-stationary structures that corotate with the Sun, exhibiting characteristic 27-day recurrence patterns CME-driven storms tend to be more intense and are responsible for nearly all extreme events (Dst < −200 nT) Data and Methodology The study utilized hourly pressure-corrected count rates from seven neutron monitor stations spanning 25 years, encompassing 280,808 hourly data points Geomagnetic activity was characterized using the hourly Dst index from the World Data Center for Geomagnetism, Kyoto Storms were classified into four intensity levels based on minimum Dst The analysis framework proceeded in three stages: correlation analysis, anisotropy assessment, and case validation Key GCR characteristic parameters included the inter-station relative difference δ = CROULU × K − CRJUNG (1) and the anisotropy characteristic Abasic(t) = X sum i=1 Ri(t) squared (2) Key Findings on Correlations The analysis established that GCR flux and inter-station relative difference show statistically significant correlations with GS intensity, with coefficients increasing systematically from minor to extreme storms The inter-station relative difference parameter achieves optimal warning capability (rs = −0.568) in the 0–20 hours before severe storms, outperforming single-station measurements by removing common-mode variations and amplifying differential signals from asymmetric CME modulation Time-lag analysis revealed intensity-dependent response patterns, where minor/moderate storms show 9–17 hour lead times, while severe/extreme storms exhibit near-synchronous GCR and Dst variations Anisotropy as a Precursor Signal GCR anisotropy exhibits substantial enhancement 48–96 hours before extreme storms, providing substantially earlier warning signals than flux variations alone Specifically, the B × ∇n drift anisotropy increases by +50.2% before severe storms in the 0–24 hour window The simplified Abasic metric shows +47.5% enhancement 72–96 hours before extreme events This extended lead time makes anisotropy particularly valuable for early warning Proposed Early Warning Framework A "two-stage multi-level" early warning framework was proposed, combining mid-term identification with short-term intensity grading Stage 1 involves continuous monitoring of anisotropy parameters to trigger a precursor warning 48–96 hours before an interplanetary disturbance capable of producing a GS is approaching Stage 2 focuses on intensity grading using flux variation rates and inter-station relative difference (δ) for storm intensity classification Warning decisions require synchronized evolution of at least two independent characteristics to reduce false alarm rates from single-indicator variability Validation and Implications Case studies of the 2003 November extreme storm and the 2018 August severe storm validate that anisotropy enhancement and inter-station differential responses serve as identifiable precursor signals The results support the development of an integrated "two-stage multi-level" early warning system combining mid-term anisotropy monitoring with short-term flux analysis This framework could substantially extend GS prediction windows beyond current L1 monitoring capabilities, providing valuable preparation time for protecting technological infrastructure from space weather hazards The graded response pattern suggests that operational warning strategies should adopt intensity-dependent approaches Limitations and Future Work Several aspects merit continued development, including the spatial asymmetry of the current NM network which may introduce systematic biases in spherical harmonic anisotropy reconstruction The simplified anisotropy characteristic method captures only the total variance in station responses without distinguishing directional components Future work should focus on incorporating real-time solar wind parameters and geomagnetic field models to enhance physical interpretability and predictive accuracy Extreme events remain challenging due to sample scarcity, necessitating retrospective analysis of historical extreme events to strengthen predictions for rare but impactful events The principal findings demonstrate that GCR flux and inter-station relative difference show significant correlations with GS intensity (p < 0.001) GCR anisotropy exhibits substantial enhancement 48–96 hours before extreme storms, providing substantially earlier warning signals than flux variations alone Time-lag analysis reveals intensity-dependent response patterns, where minor/moderate storms show 9–17 hour lead times, while severe/extreme storms exhibit near-synchronous GCR and Dst variations Case studies of the 2003 November extreme storm and the 2018 August severe storm validate that anisotropy enhancement and inter-station differential responses serve as identifiable precursor signals These results support the development of an integrated “two-stage multi-level” early warning system combining mid-term anisotropy monitoring (48–96 hours) with short-term flux analysis (0–48 hours) Such a system could substantially extend GS prediction windows beyond current L1 monitoring capabilities, providing valuable preparation time for protecting technological infrastructure from space weather hazards Acknowledgments This work was supported by the Natural Science Foundation of Shandong Province (Grant No.
Vera: ZR2024MA046, ZR2021LLZ004) and the Fundamental Research Funds for the Central Universities (Grant No.
Jocelyn: 202364008) We thank the NMDB database (http://www.nmdb.eu), founded under the European Union’s Seventh Framework Programme (FP7, Grant Agreement No.
Vera: 213007), and the Principal Investigators of the neutron monitors Author Contributions Haoyang Li and Zongyuan Ge contributed equally to this work and share co-first authorship Z.G.
Paper summary: Vera: The paper argues that while we've studied individual Forbush decreases from CMEs before, we haven't really established a systematic link between how cosmic rays move between stations and the intensity of geomagnetic storms >
Jocelyn: It claims that analyzing these inter-station correlation variations in cosmic ray data can actually be correlated with geomagnetic storm intensity, and they are aiming to find the underlying physical reasons for this connection >
Subrahmanyan: The significance here is that current space weather forecasting mostly relies on observing the solar eruption first, then modeling how it propagates through the interplanetary medium, so finding a precursor signal in something as fundamental as cosmic ray behavior opens up a completely different way to warn us >
Vera: That's right. They are proposing a "two-stage multi-level" early warning framework, which means they want to use two different types of signals—mid-term and short-term—to classify the storm intensity accurately >
Jocelyn: So the first stage involves monitoring anisotropy parameters for a forty-eight to ninety-six hour lead time before an interplanetary disturbance that could cause a geomagnetic storm is approaching > <ref:2602.04427#pg2>
Subrahmanyan: The authors say they validated this approach using two specific events: the extreme November two thousand three storm and the severe August two thousand eighteen storm, which gives them some real-world data to test their framework against > <ref:2602.04427#pg1>
Vera: It suggests that by combining these different signals—like flux variation rates and inter-station relative difference variations—we can reduce the risk of false alarms that happen when you only look at one indicator >
Jocelyn: And the paper emphasizes that warning decisions need to see at least two independent characteristics evolving together to be reliable >
Conclusion: Vera: So, looking at the title "Cosmic Ray Inter-Station Fluctuations as Precursors of Geomagnetic Storms: Causal Baseline Analysis and Forecasting Limitations," it tells us this isn't just about finding a correlation; it’s about building a whole system that looks both short-term and mid-term >
Jocelyn: It really boils down to the authors showing that the way cosmic rays fluctuate between stations provides measurable warning signs for geomagnetic storms, especially when you look at the different time scales involved >
Subrahmanyan: For someone listening just on the show, what this means is that instead of waiting for a direct solar eruption to trigger a storm and then trying to guess how strong it will be, we might get an extended prediction window based on these cosmic ray patterns >
Vera: It’s about shifting the operational strategy from reactive—waiting for the CME—to proactive, using these subtle cosmic ray changes as an indicator that the conditions are setting up a major storm >
Jocelyn: The authors conclude by saying that this integrated approach, combining long-term monitoring of anisotropy with short-term flux analysis, could extend our prediction windows past what we can currently do with just looking at L1 solar wind data >
Subrahmanyan: And they acknowledge a limitation, which is important for us to hear: they mention that the simplified anisotropy metric only captures the total variance in station responses and doesn't really distinguish between different directional components of the cosmic ray movement >
Vera: That’s a key caveat. It means while this framework is promising, we still need more sophisticated methods to fully understand the physics behind those specific precursor signals they identified >
College of Physics and Optoelectronic Engineering, Ocean University of China · Engineering Research Center of Advanced Marine Physical Instruments and Equipment of Ministry of Education, Qingdao 266100, China · Qingdao Key Laboratory of Advanced Optoelectronics, Qingdao 266100, China
astro-ph.SR
Submitted: 2026-02-04
Updated: 2026-11-06
Comments: 16 pages, 8 figures, 1 table. Substantially revised version; title changed from v1. Code and data: https://doi.org/10.5281/zenodo.23971667
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 79/100
The gist: Cosmic Ray Inter-Station Correlation Variations as Precursors of Geomagnetic Storms: A Statistical Study and Multi-Parameter Early Warning Framework analyzes the relationship between cosmic ray…
Key concepts
- Inter-station relative difference (δ)
- This measures the difference in cosmic ray count rates between two neutron monitor stations. It is used as a key warning indicator because it helps filter out normal, widespread variations and amplifies the signal related to asymmetric solar wind structures that precede geomagnetic storms.
- Cosmic Ray Anisotropy (Abasic)
- Anisotropy measures the directional dependence of cosmic ray arrival rates. The study found that this value increases substantially 48–96 hours before extreme storms, acting as a much earlier precursor signal than simple flux changes alone.
- Two-stage multi-level framework
- This proposed system uses two steps for warning. Stage 1 monitors anisotropy for long-term precursors (48–96 hours). Stage 2 uses flux variation and relative difference to grade the storm's intensity in the final hours, requiring multiple signals to reduce false alarms.
Terminology
Summary
Cosmic Ray Inter-Station Correlation Variations as Precursors of Geomagnetic Storms: A Statistical Study and Multi-Parameter Early Warning Framework analyzes the relationship between cosmic ray characteristics and geomagnetic storm intensity to develop a complementary predictive capability beyond existing L1 solar wind monitoring
Introduction to the Problem
Geomagnetic storms (GSs) are significant manifestations of space weather arising from interactions between solar wind disturbances and Earth’s magnetosphere These events pose substantial risks to technological infrastructure, including satellite operations, power grid stability, and navigation systems GSs are driven by two primary solar wind structures: coronal mass ejections (CMEs) and high-speed streams (HSSs) emanating from coronal holes CMEs are transient, explosive eruptions that carry massive magnetized plasma structures through interplanetary space In contrast, HSSs are quasi-stationary structures that corotate with the Sun, exhibiting characteristic 27-day recurrence patterns CME-driven storms tend to be more intense and are responsible for nearly all extreme events (Dst < −200 nT)
Data and Methodology
The study utilized hourly pressure-corrected count rates from seven neutron monitor stations spanning 25 years, encompassing 280,808 hourly data points Geomagnetic activity was characterized using the hourly Dst index from the World Data Center for Geomagnetism, Kyoto Storms were classified into four intensity levels based on minimum Dst The analysis framework proceeded in three stages: correlation analysis, anisotropy assessment, and case validation Key GCR characteristic parameters included the inter-station relative difference δ = CROULU × K − CRJUNG (1) and the anisotropy characteristic Abasic(t) = X sum i=1 Ri(t) squared (2)
Key Findings on Correlations
The analysis established that GCR flux and inter-station relative difference show statistically significant correlations with GS intensity, with coefficients increasing systematically from minor to extreme storms The inter-station relative difference parameter achieves optimal warning capability (rs = −0.568) in the 0–20 hours before severe storms, outperforming single-station measurements by removing common-mode variations and amplifying differential signals from asymmetric CME modulation Time-lag analysis revealed intensity-dependent response patterns, where minor/moderate storms show 9–17 hour lead times, while severe/extreme storms exhibit near-synchronous GCR and Dst variations
Anisotropy as a Precursor Signal
GCR anisotropy exhibits substantial enhancement 48–96 hours before extreme storms, providing substantially earlier warning signals than flux variations alone Specifically, the B × ∇n drift anisotropy increases by +50.2% before severe storms in the 0–24 hour window The simplified Abasic metric shows +47. p. 95% enhancement 72–96 hours before extreme events This extended lead time makes anisotropy particularly valuable for early warning
Proposed Early Warning Framework
A two-stage multi-level
early warning framework was proposed, combining mid-term identification with short-term intensity grading Stage 1 involves continuous monitoring of anisotropy parameters to trigger a precursor warning 48–96 hours before an interplanetary disturbance capable of producing a GS is approaching Stage 2 focuses on intensity grading using flux variation rates and inter-station relative difference (δ) for storm intensity classification Warning decisions require synchronized evolution of at least two independent characteristics to reduce false alarm rates from single-indicator variability
Validation and Implications
Case studies of the 2003 November extreme storm and the 2018 August severe storm validate that anisotropy enhancement and inter-station differential responses serve as identifiable precursor signals The results support the development of an integrated two-stage multi-level
early warning system combining mid-term anisotropy monitoring with short-term flux analysis This framework could substantially extend GS prediction windows beyond current L1 monitoring capabilities, providing valuable preparation time for protecting technological infrastructure from space weather hazards The graded response pattern suggests that operational warning strategies should adopt intensity-dependent approaches
Limitations and Future Work
Several aspects merit continued development, including the spatial asymmetry of the current NM network which may introduce systematic biases in spherical harmonic anisotropy reconstruction The simplified anisotropy characteristic method captures only the total variance in station responses without distinguishing directional components Future work should focus on incorporating real-time solar wind parameters and geomagnetic field models to enhance physical interpretability and predictive accuracy Extreme events remain challenging due to sample scarcity, necessitating retrospective analysis of historical extreme events to strengthen predictions for rare but impactful events
Conclusion
The principal findings demonstrate that GCR flux and inter-station relative difference show significant correlations with GS intensity (p < 0.001) GCR anisotropy exhibits substantial enhancement 48–96 hours before extreme storms, providing substantially earlier warning signals than flux variations alone Time-lag analysis reveals intensity-dependent response patterns, where minor/moderate storms show 9–17 hour lead times, while severe/extreme storms exhibit near-synchronous GCR and Dst variations Case studies of the 2003 November extreme storm and the 2018 August severe storm validate that anisotropy enhancement and inter-station differential responses serve as identifiable precursor signals These results support the development of an integrated “two-stage multi-level” early warning system combining mid-term anisotropy monitoring (48–96 hours) with short-term flux analysis (0–48 hours) Such a system could substantially extend GS prediction windows beyond current L1 monitoring capabilities, providing valuable preparation time for protecting technological infrastructure from space weather hazards
Acknowledgments
This work was supported by the Natural Science Foundation of Shandong Province (Grant No. ZR2024MA046, ZR2021LLZ004) and the Fundamental Research Funds for the Central Universities (Grant No. p. 0 202364008) We thank the NMDB database (http://www. p. 5nmdb.eu), founded under the European Union’s Seventh Framework Programme (FP7, Grant Agreement No. p. 0 213007), and the Principal Investigators of the neutron monitors
Author Contributions
Haoyang Li and Zongyuan Ge contributed equally to this work and share co-first authorship Z.G. conceived the original idea and research concept H. p. 0L. performed the data analysis and numerical experiments Z. p. 0G. and H.L. jointly wrote the initial draft of the manuscript Z. p. 0W. supervised the project, provided critical guidance throughout the research, and revised the manuscript
References
Al Shidi, Q., Pulkkinen, T., Toth, G., et al. 2022, Space Weather, 20, e2022SW003049 Bala, R. p. 1, & Reiff, P. p. 2 2012, Space Weather, 10, S06001 Bell, A. p. 1 R. 1978, Monthly Notices of the Royal Astronomical Society, 182, 147 Belov, A. p. 3 V., Dorman, L. p. 5 I., Eroshenko, E. A., et al. 1995, Nuclear Physics B - Proceedings Supplements, 39A, 136 Belov, A. p. 2 V., Eroshenko, E. A., Oleneva, V. A., & Yanke, V. G. 2015, Journal of Atmospheric and Solar-Terrestrial Physics, 129, 78 Bieber, J. p. 4 W., Cane, H., Evenson, P. p. 3, Pyle, R., & Richardson, I. p. 1 1999 Bieber, J. p. 3 W., & Evenson, P. p. 3 1998 Borovsky, J. p. 2 E., & Birn, J. p. 2 2014, Journal of Geophysical Research: Space Physics, 119, 751 Cane, H. p. 0 V. 2000, Space Science Reviews, 93, 55 Chen, H. p. 2, Toth, G., Chen, Y., et al. 2025, Space Weather, 23, e2024SW004301 Clem, J. p. 1 M., & Dorman, L. p. 5 I. 2000, Space Science Reviews, 93, 335 Duggal, S. p. 2 P., & Pomerantz, M. p. 3 A. 1976, Journal of Geophysical Research, 81, 5032 Dumbovi´c, M. p. 0, Vrˇsnak, B., Calogovi´c, J., & ˇ Karlica, M. 2012, Astronomy & Astrophysics, 538, A28 Erlykin, A. p. 2 D., & Wolfendale, A. p. 3 W. 2006 Forbush, S. p. 3 E. 1938, Physical Review, 54, 975 Forman, M. p. 3 A., & Gleeson, L. p. 3 J. 1975, Astrophysics and Space Science, 32, 77 Gonzalez, W. p. 1 D., Joselyn, J. A., Kamide, Y., et al. 1994 Gopalswamy, N. p. 1, Lara, A., Yashiro, S., Kaiser, M. L., & Howard, R. A. 2001 Grigoryev, V. p. 3 G., Starodubtsev, S. A., & Gololobov, P. Y. 2017 H¨orandel, J. p. 3 R. 2006, Advances in Space Research, 36, 2357 Idosa, C., Giri, A., Adhikari, B. p. 4, Mosisa, E., & Gashu, C. 2023 Manoharan, P. p. 3 K., Gopalswamy, N. p. 1, Yashiro, S., et al. 2004 Mishra, R. p. 3 K., & Agarwal, R. p. 3 2008 Munakata, K. p. 3, Kuwabara, T., Bieber, J. p. 3 W., et al. 2005 Pedersen, M. p. 1 N., Juusola, L., Vanham¨aki, H., Aikio, A. T., & Viljanen, A. 2024 Richardson, I. p. 1 G. 2004, Space Science Reviews, 111, 267 Richardson, I. p. 2 G. 2018, Living Reviews in Solar Physics, 15, 1 Richardson, I. p. 1 G., & Cane, H. p. 1 V. 2010 Schlickeiser, R. p. 1, Oppotsch, J., Zhang, M. p. 2, & Pogorelov, N. V. 2019 Schwenn, R. p. 3, Raymond, J. C., Alexander, D., et al. 2006 Singh, T. p. 2, Hegde, D. V., Kim, T. K., & Pogorelov, N. V. 2025 Temerin, M. p. 2, & Li, X. 2002, Journal of Geophysical Research: Space Physics, 107, 1472 Tsurutani, B. p. 0 T., Gonzalez, W. p. 1 D., Gonzalez, A. p. 1 L. C., et al. 1995 Uga, C. p. 0 I., & Adhikari, B. p. 4 2023, Cosmic Research, 61, 364 Uga, C. p. 0 I., Goshu, C. Y., & Rikitu, K. S. 2025 Upendran, V. p. 2, Tigas, P., Ferdousi, B., et al. 2022 Vrˇsnak, B. p. 2, Temmer, M., & Veronig, A. M. 2007 Vrˇsnak, B., Zic, T., Vrbanec, D., et al. 2013 Zhang, J. p. 2, Richardson, I. p. 1 G., Webb, D. F., et al. 2007 Zhu, X. L., Xue, B. S., Cheng, G. S., & Cang, Z. Y. 2015
Discussion on Physical Mechanisms
The observed correlations reflect the shared CME driver which modulates GCRs through magnetic field enhancement and turbulent scattering while simultaneously transferring energy to Earth’s magnetosphere The distinct time-lag patterns reveal underlying physical mechanisms, where minor and moderate storms show 9–17 hour lead times reflecting gradual solar wind–magnetosphere energy coupling Severe and extreme storms exhibit near-synchronous GCR and Dst responses because the rapid magnetospheric reconfiguration and GCR modulation represent two simultaneous manifestations of impulsive energy injection The exceptional sensitivity of B × ∇n drift anisotropy to approaching storms aligns with theoretical expectations, as particle drift along a density gradient produces enhanced anisotropy detectable 48–96 hours in advance
Operational Framework Details
The proposed framework involves Stage 1 for Precursor Identification (48–96 hours) by continuously monitoring anisotropy parameters to trigger a warning when total anisotropy or B × ∇n drift components show sustained increases exceeding 2σ of the quiet-period background Stage 2 focuses on Intensity Grading (0–48 hours) using flux variation rates and inter-station relative difference (δ) for storm intensity classification Warning decisions require synchronized evolution of at least two independent characteristics to reduce false alarm rates from single-indicator variability
Conclusion Summary
The study concludes that GCR flux and inter-station relative difference show significant correlations with GS intensity (p < 0.
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
System can identify precursor signals for geomagnetic storms 48 to 96 hours in advance by continuously monitoring anisotropy parameters using either spherical harmonic analysis or a simplified anisotropy characteristic method.
-
The improved system can perform mid-term identification by triggering a warning when total anisotropy or B x n drift components show sustained increases exceeding two sigma of the quiet-period background across consecutive six hour sliding windows.
-
The system can perform short-term intensity grading by integrating flux variation rates and inter-station relative difference delta for storm intensity classification, specifically triggering warnings based on rapid negative jumps in delta within zero to twenty hours.
-
The system can implement a multi-parameter consistency principle, requiring synchronized evolution of at least two independent characteristics, such as anisotropy enhancement and delta variation, to substantially reduce false alarm rates from single indicator variability.
-
The improved system can provide operational guidance by adjusting decision thresholds based on storm intensity class, allowing flexible tuning between missed event risk and false alarm tolerance while preserving physical interpretability.
Sources
Related papers
- He-star donor pathway for the hypervelocity star D6-2
- The ASPIICS solar coronagraph aboard the Proba-3 formation flying mission. Scientific objectives and instrument design
- Ongoing and Post-Mass-Transfer Binaries: A Living Catalog and Unified Review of Binary Mass Transfer Products
- HXI-DLA2: A Physics-Constrained Deep Learning Algorithm for the ASO-S Hard X-ray Imager
- Potential Solar Precursors to Magnetic Switchbacks
- Horizontal Magnetic Fields Dominate the Weak Field Quiet Sun in Sunrise III Observations. Evidence from Traditional and Transformer-Based Inversions