Use reverse-forward declustering discrepancy as a diagnostic

Establish whether the discrepancy between forward nearest-neighbor and reverse nearest-neighbor declustering can serve as an empirical diagnostic of preparatory seismic organization.

Background

The study finds that reverse nearest-neighbor declustering identifies more foreshocks than standard forward-in-time nearest-neighbor declustering, particularly for small foreshocks and mainshocks. The authors emphasize that this difference may arise from the temporal asymmetry of the metric rather than from a genuine preparatory physical mechanism.

Accordingly, the paper leaves unresolved whether the contrast between forward and reverse declustering results contains information about physical preparation beyond methodological bias. Testing this possibility would require applying the comparison to additional tectonic regions and relating the discrepancy to independently established indicators of preparatory seismic organization.

References

Future work should extend this comparison to additional tectonic regions, explore adaptive combinations of NN and rNN metrics, and test whether the discrepancy between forward and reverse declustering can be used as an empirical diagnostic of preparatory seismic organization.

— Is the Number of Foreshocks Underestimated? The Role of Declustering Symmetry  (2610.13147 - Prabowo et al., 9 Oct 2026) in Section 6, Conclusions