Reports of disturbances foreshadowing earthquakes are common, and have been disproven quickly in nearly all cases.  However, one recent claim by Kosuke Heki survived the review process to appear in the journal Geophysical Research Letters, and even was featured in the linked research spotlight.

At the American Geophysical Union meeting in San Francisco this week, at the end of the highlighted Bowie Lecture, Dr. Heki said it again and gave more details.  Not only the deadly 2011 Tohoku earthquake, but also several other huge earthquakes appear to show more charged particles above the ground in the roughly one hour before the earthquakes struck.  The signal for Japan was claimed to be local to the area of the Japanese earthquake, as indicated in this figure.

Another study claimed the ionospheric disturbance is just a solar storm, and it is global rather than local to part of Japan.  And that the signals before most of the other large earthquakes are also solar storms.  Which begs the question of whether solar storms might trigger earthquakes, a possibility considered even more unlikely by scientists.

How such a precursor might be generated is not known except in outlandish theories, and most scientists remain skeptical that the signal is real.  However, the claim is specific and testable, so there is on the table a serious idea from a serious scientist for once in earthquake prediction.