On the correlation between solar activity and large earthquakes worldwide

Large earthquakes occurring worldwide have long been recognised to be non Poisson distributed, so involving some large scale correlation mechanism, which could be internal or external to the Earth. Till now, no statistically significant correlation of the global seismicity with one of the possible mechanisms has been demonstrated yet. In this paper, we analyse 20 years of proton density and velocity data, as reported by the ISC-GEM catalogue. We found clear correlation between proton density and the occurrence of large earthquakes (M>5.8), with a time shift of one day. The significance of such correlation is very high, with probability to be wrong less than 10-5. The correlation increases with the magnitude threshold of the seismic catalogue. A tentative model explaining such a correlation is also proposed, in terms of the reverse piezoelectric effect induced by the applied electric field. This result opens new perspectives in seismological interpretations, as well as in earthquake forecast.

💡 Research Summary

The paper investigates whether solar activity, as measured by proton density and velocity in the solar wind, is correlated with the occurrence of large earthquakes worldwide. Using two extensive data sets—(i) the ISC‑GEM global earthquake catalogue (M ≥ 5.8) covering 1996‑2015, and (ii) hourly proton measurements from the SOHO satellite at the L1 point—the authors construct daily averages of four proton‑related variables: density (ρ), velocity (v), flux (ρv), and dynamic pressure (ρv²/2). For each variable they define a non‑dimensional average and scan a series of thresholds (V_T) from the mean up to the maximum, stepping by 0.01. Six temporal conditions are examined (e.g., “one day after the variable falls below the threshold”). For each condition and threshold they count the number of days D_C that satisfy the condition and the number of earthquakes E_C occurring on those days, then compute a relative event rate R = (E_C/D_C)/(E/D), where E and D are the total events and days in the full catalogue.

To assess statistical significance, the authors generate 100,000 synthetic earthquake catalogs by randomly permuting the observed inter‑event intervals, thereby preserving the empirical survival function (the distribution of waiting times) while destroying any true temporal association with the solar data. For each synthetic catalog they recompute R under the same conditions, producing a distribution of R_rand values. An observed R is declared significant only if it exceeds all R_rand values, corresponding to a false‑positive probability < 10⁻⁵ (confidence > 99.999 %). This bootstrap approach explicitly accounts for the known non‑Poissonian nature of global seismicity.

The analysis reveals that only the proton density (ρ) and the derived flux (ρv) show a statistically significant increase in earthquake rate when the variable drops below a threshold and the following day is examined (condition “1Dy bT”). The effect is strongest for density thresholds corresponding to 12.7–15.9 cm⁻² (V_step ≈ 0.31–0.39). Under these conditions, the relative rate R rises to roughly 1.5–2.0, indicating a 50–100 % increase over the background. Moreover, when the magnitude cut‑off of the seismic catalog is raised (M ≥ 6.0, 6.5, 7.0), the peak R becomes larger, suggesting that the correlation is more pronounced for the largest events. No significant relationship is found for proton velocity or dynamic pressure.

In the discussion, the authors propose a qualitative physical mechanism: high proton density in the ionosphere raises the electric potential of the upper atmosphere relative to the Earth’s surface. This potential difference could drive currents along highly conductive fault zones, producing a reverse piezoelectric effect in quartz‑rich rocks. The resulting strain pulse would add to the tectonic loading, potentially triggering rupture. They link this hypothesis to reported phenomena such as earthquake lights and pre‑seismic radio emissions, arguing that these electromagnetic signatures could be manifestations of the same ionospheric‑earth coupling.

The paper concludes that a robust, statistically significant correlation exists between global large‑earthquake occurrence and solar‑wind proton density, and that this finding may open new avenues for seismological research and earthquake forecasting. However, the authors acknowledge that a quantitative model of the proposed mechanism is beyond the scope of the present work.

Critical appraisal: While the statistical methodology is rigorous in its use of bootstrapping and a very high confidence threshold, several limitations deserve attention. First, the seismic catalog is used in its raw, non‑declustered form; aftershocks are therefore treated as independent events, which can inflate apparent correlations. Second, the proton density series itself exhibits non‑white spectral peaks (notably near 27 days, the solar rotation period), raising the possibility that the observed correlation reflects a common periodic driver rather than a causal link. Third, the bootstrap preserves the inter‑event time distribution but does not model the physical stress accumulation on faults, so the null hypothesis may be overly simplistic. Fourth, the proposed reverse piezoelectric mechanism is presented qualitatively without any quantitative estimates of the required electric fields, current densities, or stress changes, making it difficult to evaluate plausibility. Finally, the study focuses exclusively on one solar‑wind parameter; inclusion of other solar activity indices (e.g., sunspot number, solar flare flux) could help disentangle whether proton density is the primary driver or a proxy for a broader solar‑terrestrial coupling.

Future work should address these issues by (i) declustering the seismic catalog to isolate independent mainshocks, (ii) performing spectral coherence analysis between proton density and seismicity to test for shared periodicities, (iii) incorporating additional solar and geomagnetic observables, and (iv) developing a physics‑based model that links ionospheric electric potentials to fault‑scale stress perturbations, complete with realistic conductivity and piezoelectric parameters. Only with such multidisciplinary validation can the claimed solar‑earth connection move from statistical association to a credible causal hypothesis.