Revelation of early detection of co-seismic ionospheric perturbations in GPS-TEC from realistic modelling approach: Case study

Abstract

GPS-derived Total Electron Content (TEC) is an integrated quantity; hence it is difficult to relate the detection of ionospheric perturbations in TEC to a precise altitude. As TEC is weighted by the maximum ionospheric density, the corresponding altitude (hmF2) is, generally, assumed as the perturbation detection altitude. To investigate the validity of this assumption in detail, we conduct an accurate analysis of the GPS-TEC measured early ionospheric signatures related to the vertical surface displacement of the Mw 7.4 Sanriku-Oki earthquake (Sanriku-Oki Tohoku foreshock). Using 3D acoustic ray tracing model to describe the evolution of the propagating seismo-acoustic wave in space and time, we demonstrate how to infer the detection altitude of these early signatures in TEC. We determine that the signatures can be detected at altitudes up to ~130 km below the hmF2. This peculiar behaviour is attributed to the satellite line of sight (LOS) geometry and station location with respect to the source, which allows one to sound the co-seismic ionospheric signatures directly above the rupture area. We show that the early onset times correspond to crossing of the LOS with the acoustic wavefront at lower ionospheric altitudes. To support the proposed approach, we further reconstruct the seismo-acoustic induced ionospheric signatures for a moving satellite in the presence of a geomagnetic field. Both the 3D acoustic ray tracing model and the synthetic waveforms from the 3D coupled model substantiate the observed onset time of the ionospheric signatures. Moreover, our simple 3D acoustic ray tracing approach allows one to extend this analysis to azimuths different than that of the station-source line.

Ionospheric GNSS Imagery of Seismic Source: Possibilities, Difficulties, and Challenges

“…Starting from 470 s, the TEC on the east‐northeast from the epicenter exceeds the background level, so that we can affirm reliable detection of a CID. We emphasize that this area of the first enhanced TEC corresponds to the uplift area as was estimated by Thomas et al () and as shown by white cross in Figure a. This first CID seems to further move southward while simultaneously extending westward.…”

“…For the 2011 Sanriku‐oki earthquake, by using only data of satellite G07 Thomas et al () found extremely early first CID arrivals and explained them by a particular geometry of sounding at low‐elevation angles. Thomas et al () found that the first ionospheric detection took place at around 130 km of altitude, which is much below the ionization maximum of 275 km as follows from the International Reference Ionosphere (IRI) model (Bilitza et al, ; Figure a). Our results here confirm that for both G07 and G10 measurements the first CID occur 470–480 s after the earthquake (Figures b–d).…”

“…(a) General information about the Sanriku‐oki earthquake of 9 March 2011. The epicenter is shown by red star; the position of the maximum of the coseismic crustal uplift is depicted by white cross (data from Thomas et al, ). The magnitude of the coseismic slip is shown by colored dots around the epicentral area, and the corresponding color scale is shown below the panel (the data are from the U.S. Geological Survey, http://earthquake.usgs.gov).…”

Ionospheric Detection of Natural Hazards

“…The location of these two TEC enhancements corresponded to the location of two segments of coseismic crustal slip as it was shown by seismologists (e.g., Simons, ; Bletery et al, ; Figure a). More recently, Thomas et al () and Astafyeva and Shults () demonstrated that the GNSS seismo‐ionospheric imagery can work for moderate earthquakes as well. Astafyeva and Shults () also noted that lower resolution data sampling, such as 15 or 30 s, which so far is a standard resolution for GNSS data, will, most likely, not work efficiently.…”

“…Within the past two decades, numerous simulation techniques have been developed. For instance, ray tracing is the simplest, although not the most precise, way to reproduce near‐field ionospheric signatures of earthquakes and volcanic eruptions (Calais et al, ; Dautermann, Calais, & Mattioli, ; Heki et al, ; Heki & Ping, ; Rolland et al, ; Thomas et al, ). To model CSID propagation in the far‐field and the ionospheric signatures of Rayleigh surface waves and tsunamis, normal mode summation technique shows promising results (Artru et al, ; Artru et al, ; Coïsson et al, ; Lognonné et al, ; Rakoto et al, ; Rolland, Lognonné, & Munekane, ).…”