IASPEI - Picture Gallery

IUGG XXIV General Assembly July 2-13, 2007 Perugia, Italy
(S) - IASPEI - International Association of Seismology and Physics of the Earth's
Interior
JSS008 Oral Presentation 1980
Comments on seismomagnetic effects from the long-awaited 28
September 2004 M6.0 Parkfield earthquake by M.J.S. Johnston, Y. Sasai,
G.D. Egbert, R.J. Miller
Prof. Panayiotis Varotsos
Physics Department University of Athens IASPEI
Seiya Uyeda
Johnston et al. [1] list four criticisms on Seismic Electric Signals (SES) [2]. (1) There are no similar coseismic
signals observed when the primary earthquake energy is released. Actually, signals at the time
of EQ observed routinely in Greece, Japan, Indonesia and even at Parkfield were all co-seismic wave.
However, this does not negate SES. SES generation takes place during the slow stress increase and has
little to do with sudden stress release. Why stress release does not generate observable signals is a
separate question. (2) No clear physical explanation exists describing how the SES signals can relate to
EQs occurring sometimes hundred of kilometers away. This statement is untrue. Structural
inhomogeneity may be the key: The EQ fault may be orders of magnitude more conductive than the
surrounding, providing a conductive path for SES[2]. This explains why SES reveals the so called
selectivity. In other words, unless the observation point is at a sensitive site, SES can not be observed.
(3) no independent data (strain, seismic, pore pressure etc) exists that supports the proposed
earthquake/SES relationship. Again, these authors do not seem to understand the SES generation
mechanism well. SES is a critical phenomenon in the sense that it is generated when the stress reaches
a certain level, requiring no sudden changes. Some geophysical parameters may coincidently show
anomalies. In fact, recent studies by natural time analysis have shown that pre-mainshock seismicity
exhibits important changes just after SES, but that is not a prerequisite for SES generation. (4) The SES
signals have the form expected from rectification/saturation effects of local radio transmissions (Pham
et al., 1998). However, Pham et al.[3] argument proves nothing[4]. They observed SES-looking noises
in Greece and stated that what we observe were all such noises. We also observed them but have
discarded by applying our noise discriminating criteria. Johnston et al. did not see SES at Parkfield EQ.
Their results from a single station alone do not prove that SES did not exist. As stated above, SES are
recorded only at a sensitive site, whereas they have not made any selective site search. Thus, it seems
much too premature to conclude either Parkfield EQ was not preceded by SES or useful prediction of
damaging earthquakes seems unlikely using electromagnetic data. [1] M.J.S. Johnston et al., BSSA. 96,
no.4B, S206-S220 (2006). [2] P. Varotsos, The Physics of Seismic Electric Signals, TerraPub, Tokyo,
338pp (2005). [3] V.N. Pham et al., GRL. 25, 2229-2232 (1998). [4] N. Sarlis et al., GRL. 26, 3245-3248
(1999).
Keywords: seismic electric signals, parkfield