JpGU-AGU Joint Meeting 2020

Presentation information

[E] Poster

S (Solid Earth Sciences ) » S-SS Seismology

[S-SS07] Earthquake prediction and forecast+pre-earthquake geoscientific changes(TBD)

convener:Masayuki Obayashi(Japan Agency for Marine-Earth Science and Technology), Toshitaka Baba(Graduate School of Science and Technology, Tokushima University), Chi-Yu King(Retired)

[SSS07-P05] Ionosphere Magnetic Field Precursor Anomalies Extraction Associated with 2016 Ecuador Earthquake using NMF

*Mengxuan Fan1, Kaiguang Zhu1, Xiaodan He1, Kaiyan Li1, Dedalo Marchetti2, Chengquan Chi1, ZINING YU1 (1.Jilin University, 2.National Institute of Geophysics and Volcanology)

Keywords:Eearthquake, Satellites magnetic field, non-negative matrix factorization, lithosphere atmosphere ionosphere coupling

Some ionosphere anomalies are found in satellite observations during the preparation phase of earthquakes. In previous earthquake studies, the magnetic field data under strong geomagnetic activity are usually deleted to avoid the influence of the geomagnetic activity, therefore the dataset are reduced. In this paper, we decompose the Swarm satellite time series magnetic field data under all geomagnetic conditions and all local time by Nonnegative Matrix Factorization to study the April 16, 2016 Mw7.8 Ecuador earthquake. The data are decomposed into basic vectors and their corresponding weights, and we find that the energy and entropy of component H1, one of the weight components, are more concentrated inside the earthquake sensitive area than other weight components. The cumulative number of anomalous tracks with anomalies inside the earthquake sensitive area evaluated by component H1 shows accelerated growth before the earthquake and recovers to linear growth after, which obey the power law behavior of a critical system, while for other components H2 and H3 show linear growth before and after the earthquake. Furthermore, the cumulative anomalies of component H1 also correspond to two previous studies which also use the magnetic field data of Swarm satellite, but removed the data or component influenced by the geomagnetic activity. And the test on the anomalous track percentages versus different geomagnetic levels indicates the cumulative anomalies of component H1 are hardly affected by geomagnetic activity. After a confutation analysis of four random low seismicity area cases and two 90 days low seismicity period cases, the cumulative anomalies of component H1 are also confirmed as not obtained by simple chance. Moreover, we find the cumulative result of component H1 has similar growth to the cumulative Benioff strain S around the Ecuador earthquake epicenter, that suggests that the ionosphere magnetic anomalies are associated with the lithosphere activities which are related to the Ecuador earthquake and agrees with the Lithosphere Atmosphere Ionosphere Coupling.