Research on seismo-ionospheric anomalies using artificial neural network

Presently the discovery and research on the disturbances in the ionosphere before earthquake has become increasingly hot research topic. In this paper, the ionospheric occultation data of the COSMIC was used for a detailed study on the ionospheric respond caused by 7.9 magnitude earthquake occurred in Chile on November 14, 2007 and 7.9 magnitude earthquake occurred in Tonga on March 19, 2009. The paper used interpolation techniques of BP neural network, and analyzed the abnormal changes of the F2-layer peak electron density, NmF2 and F2-peak height, hmF2, over the region around the epicenter. The effective experimental results show that NmF2 and hmF2 in ionosphere apparent negative anomaly phenomenon over the region around the epicenter 5 days before the earthquake. The occurrence of these anomalies is most likely from the imminent earthquake.