Reported geomagnetic and ionospheric precursors to earthquakes: Summary, reanalysis, and implications for short-term prediction

Earthquakes are one of the most devastating natural phenomena on earth, causing high deaths tolls and large financial losses each year. If precursory signals could be regularly and reliably identified, then the hazardous effects of earthquakes might be mitigated. Unfortunately, it is not at all clear that short-term earthquake prediction is either possible or practical, and the entire subject remains controversial. Still, many claims of successful earthquake precursor observations have been published, and among these are reports of geomagnetic and ionospheric anomalies prior to earthquake occurrence. Given the importance of earthquake prediction, reports of earthquake precursors need to be analyzed and checked for reliability and reproducibility. We have done this for numerous such reports, including the Loma Prieta, Guam, Hector Mine, Tohoku, and L'Aquila earthquakes. We have found that these reported earthquake precursors: 1) often lack time series observations from long before and long after the earthquakes and near and far from the earthquakes, 2) are not statistically correlated with the earthquakes and do not relate to the earthquake source mechanisms, 3) are not followed by similar, but much larger, signals during the subsequent earthquake when the primary energy release occurs, 4) are nonuniform in that they occur at different spatial and temporal regimes relative to the earthquakes and with different magnitudes and frequencies, and 5) can often be explained by other non-earthquake related mechanisms or normal geomagnetic activity. Thus we conclude that these reported precursors could not be used to predict the time or location of the earthquakes. Based on our findings, we suggest a protocol for examining precursory reports, something that will help guide future research in this area.