Predicting Earthquakes Using GNSS Data Possible Says Japanese Scientists

Earthquake GNSSIs it really possible to predict the occurrence of earthquake? Can an earthquake be predicted before it can wreck havoc? The U.S. Geological Survey (USGS) had this to say when asked “Can you predict earthquakes?”

“No. Neither the USGS nor any other scientists have ever predicted a major earthquake. They do not know how, and they do not expect to know how any time in the foreseeable future. However based on scientific data, probabilities can be calculated for potential future earthquakes. For example, scientists estimate that over the next 30 years the probability of a major EQ occurring in the San Francisco Bay area is 67% and 60% in Southern California. The USGS focuses their efforts on the long-term mitigation of earthquake hazards by helping to improve the safety of structures, rather than by trying to accomplish short-term predictions.”

So the answer to the question is NO. There is no way an earthquake can be predicted, until this new study conducted by experts of the Japan Earthquake Science Exploration Agency (JESEA) said otherwise. Results of the scientific study showed that Global Navigation Satellite System (GNSS) signals can effectively be used as a means of earthquake prediction.

The paper aimed to validate the use of Global Navigation Satellite System (GNSS) signals prior to the occurrence of earthquakes as a means of earthquake prediction, by referring to the Great East Japan Earthquake. The Great Earthquake occurred on March 11, 2011 with Richter Scale M 9.0 and it killed about 18,000 people mainly due to the subsequent Tsunami. The validation was based on data analysis of GNSS signals at GNSS-based Control Stations, about 1,200 of which have been installed all over Japan by Geospatial Information Authority (GSI) since 1996.

The objective of the study was to detect signals before the Great Earthquake using various indicators computed from the daily GNSS data, with a view to their being used in future for the prediction of earthquakes. The study succeeded to detect several pre-signals six months, five months, one month and three days before the Great Earthquake in the daily data, averaged weekly data, weekly maximum deviations, height changes over two years and/or accumulated changes.

The most important finding in this paper was that the detection of pre-signals as well as the pre-slips in the three days prior to the Great Earthquake could have led to the prediction of the forthcoming seismic event, when it has previously been claimed to be impossible to do so with existing measuring techniques.

Consideration

The validation study was made not in advance of, but after the Great Earthquake. Therefore, there is no assurance that it would be possible to apply the same method in advance for predicting other earthquakes, as this case might be a very special one. However, one can at least obtain a hint of how GNSS data could be useful for predicting earthquakes based on various indicators computed from GNSS data.

The following problems were identified for future development of the prediction method.

1) The location and the intensity could be predicted from GNSS data as indicated in this paper, but the exact date of earthquake occurrence cannot be predicted well. The calm period after several pre-signals cannot be estimated. In this case, there was a period of about 6 months after the first pre-signal before the Great Earthquake.

2) It would be possible to determine the best fit indicator to detect critical areas in advance for considering or estimating the direction of deformation by an earthquake. The validation in this paper showed the best fit to the accumulated Y values correlated well with the movement of the Pacific coast at the Tohoku District. In the case of the different types of earthquakes and different crustal conditions, the best fit indicator would be different.

3) The pre-slip was detected after the Earthquake as the GNSS data was only obtained after the event. However, it is difficult to obtain near real time GNSS data (called R3 data), because GSI does not release such data to the private sector. Open data but with two-week delay will be too late to detect the pre-slip.

4) The long term indicators in this paper are based on a two year span for convenience, but it would be better to determine the averaged normal trend curve for several normal years.

Conclusion

It is possible to detect presignals and pre-slips with GNSS data before an earthquake as demonstrated from the data acquired prior to the Great Earthquake. This is a new development in earthquake prediction. Not only short term indicators, but also long term indicators are necessary to detect pre-signals. Simultaneous dramatic changes at multiple points enable the identification of pre-signals.

These pre-signals were detected not only over a narrow area, but also a much wider area in case of this huge earthquake. Accumulated daily changes of Y over two years were the best fit indicator to predict the most affected areas in this validation study.

Homepage: http://terras.gsi.go.jp/gps/gpsbased_ control_station.html (Japanese)
Source: http://mycoordinates.org