This event occurred on 5 February 2016 at 19:57:27 UTC east of Tainan in southern Taiwan, with magnitude 6.4, as the result of oblique thrust faulting. At the location of the EQ (22.938°N, 120.601°E) the two plates converge in a northwest-southeast direction at a velocity of about 80mm/yr [Hsu et al., 2009; Lin et al., 2010].


Figure 1: Location of the EQ epicenter and Dobrovolsky area.


From the focal mechanism of the main shock (oblique thrust with left-lateral slip component) and preliminary results of the aftershocks' distribution, the causative fault appears to be a geological unknown blind structure, probably located below the décollement of the regional fold-thrust belt, which is estimated at 8-10 km of depth. The earthquake started at about 16-18 km deep on a fault trending WNW-ESE (az:280), with a very gentle shallow dip angle of 15 degrees to the north (Report by Jian-Cheng Lee- Institute of Earth Sciences, Academia Sinica, Taipei).

Most of the aftershocks are not located on the above suspected earthquake causative fault. Instead, a large part of them lies on a cluster situated 20-30 km depth and 20 km west of main shock, below the causative fault. From the observations of the wave forms graphs, there is a strong speculation that it was a doublet earthquake with the 2nd mainshock occurring near this cluster of aftershocks, about 4 seconds following the first mainshock.




  2. Accelerated Moment Release (AMR) & Revised-AMR

The Catalogue data analysis started from January 2011 with a maximum distance of around 560 km from the epicentre. Figure 2 shows results from AMR and R-AMR analyses.

Seismicity of the considered areais characterized by many clusters of events around some large earthquakes inside the DbA of this case study. They appear as jumps especially in the first part ofthe reduced cumulative strain (Figure 2,left). However, apart from that, no acceleration exists before the mainshock.



Figure 2: AMR (left) and R-AMR (right) results.




Two different approaches were developed to search EQ precursor using geomagnetic Swarm data: MASS method and Wavelet method. Both analyses are based on Level 1B MAGxLR Swarm product.


  1. MASS algorithm (Magnetic Swarm anomaly detection by Spline analysis)

The algorithm MASS (Magnetic Swarm anomaly detection by Spline analysis)was applied to M6.4 Taiwan 2016 earthquake with different thresholds, while the moving window was fixed at 3.0°. The algorithm analyses all tracks in DbA one month before and one after the EQ. The tracks are marked as “anomaly” only if the centre of the moving window is in DbA and if geomagnetic conditions are quiet.

Figure 3 shows an example of anomalous track detected by MASS method for Taiwan EQ., while Figure 4 represents the cumulative number of anomaly tracks detected by MASS in the studied time range. No anomalies have been found for the scalar intensity F of magnetic field.



Figure 3:Example of track for Satellite C (February 6, 2016).



Figure 4: Cumulative number of anomaly detected by MASS one month before and one month after the M6.4 5 February 2016 Taiwan EQ. Threshold is kt = 2.0, the anomalies are selected only with geomagnetic quiet time (|Dst| ≤ 20 nT and ap≤ 10 nT). No anomaly tracks are found in scalar intensity “F” of magnetic field.


  1. Wavelet Analysis

The Wavelet spectral analysis has evidenced the existence (in general) of anomalous families, each characterized by some features that altogether do not clarify whether they are linked to LAIC or not.

Figure 5 shows an example of long-life anomaly detected by wavelet analysis: it involves almost the whole track passing very close to the epicentre. Anyway, it is true that something similar is also visible in Figure 6 where the track is practically at the same longitude, few days after the mainshock.



Figure 5: A long-life anomaly detected by wavelet analysis passing very close to the epicenter.


Figure 6: Anomaly track similar to that shown in Fig. 5 detected at the same longitude but few days before the EQ.

Even in this case, the wavelet analysis did not produce conclusive results so the adoption of some other more complex scheme is probably required to analyze this event.



  1. SWARM – IONOSPHERE from Satellite

Satellite-based data for the ionospheric characterization of the EQ-related events are mainly those referred to the LP (Langmuir Probe) instrument aboard the SWARM satellites. The electron density Ne is the relevant parameter used for the ionospheric characterization in the frame of SAFE project. Two different methods were developed to analyze Swarm ionospheric data: NeSTAD and NeLOG.


  1. Method I: NeSTAD

No tagged anomalies were found for this event by means of the tagging criteria applied on the NeSTAD anomaly parameters.


  1. Method II: NeLOG

The automatic search NeLOG of ionosphere electron density anomaly from Swarm data is applied to M6.4 Gibraltar case study. A sample is classified as “anomaly” if the residual value exceeds a threshold kt=2.5 of the RMS in the residuals from the fit. A track is selected as “anomaly” if it has more than 10 anomaly samples in DbA and if geomagnetic indices are |Dst|<20 nT and ap<10 nT. Tracks are selected within a mean local time between 22 and 6.

Figure 7 shows an example of anomalous track detected by NeLOG, while Figure 8 reports the cumulative number of anomaly tracks one month before and one month after the event.


Figure 7: Example of an anomaly track detected by NeLOG for Taiwan EQ (25 January 2016).




Figure 8:Cumulative number of anomaly tracks detected by NeLOG one month before and one month after M6.4 Taiwan EQ. Threshold is kt = 2.5, the anomalies are selected only with geomagnetic quiet time (|Dst| ≤ 20 nT and ap≤ 10 nT) and in night time (22 ≤ LT ≤ 6).

The cumulate plot in Figure 8 is compatible with S-shape behavior; in this case the system seems to be compatible with a critical point system: the number of the anomaly tracks increases toward the date of the EQ, and decreases after the event.


  1. IONOSPHERE Ground-based

Ionosondes & GNSS

For this event, no suitable Ionosonde data were available. For GNSS analysis, the period before the event was characterized by high geomagnetic activity and not useful for the SAFE scope.



The M6.4 Taiwan EQ was analyzed using both Swarm geomagnetic and ionospheric data searching for earthquake-related anomalies in the frame of LAIC theory.

Considering NeLOG method, the obtained value for R2 indicates that the cumulative number of anomaly tracks is not compatible with a linear fit and this could indicate a possible correlation with the EQ.

In addition, the values obtained for R2, nEQ and C factors in MASS analysis seem to suggest a possible correlation between the occurrence of anomalies and LAIC effects.

On the contrary, NeSTAD method did not found any anomalies for this event.