[G01-P] Poster

[G01-P-01] Estimation of Post Seismic Deformation Model Using Monte Carlo Method

^○Takayuki Miyazaki (GSI of Japan, Ibaraki, Japan)

The one of latest Global Geodetic Reference Frame (GGRF), ITRF2014 has been available since Jun. 2016. In ITRF2014, Post Seismic Deformation (PSD) models were newly introduced to improve accuracy of station coordinates impacted by major earthquake. PSD models will improve accuracy of coordinates of Japanese domestic GNSS observation network (GEONET) subject to active crustal deformation too. In this paper, we has developed Bayesian estimation method that based on ensemble Marcov Chain Monte Carlo (MCMC) method to make PSD models of GEONET station. This method can estimate PSD model stably avoiding local solution although PSD models includes non-linear functions. We compared estimated PSD model and that of ITRF2014 products in TSKB station, which is one of IGS GNSS station. As a result, we acquired s.d. of estimated PSD model and observation (E, N, U) = (10.1, 16.9, 1.7) mm. It is shown that estimated PSD model sufficiently explain observation data. We also evaluated prediction power of the PSD model in Miyako station, which has been impacted by the 2011 off the Pacific coast of Tohoku Earthquake. We estimated the PSD model of Miyako station with coordinate time series spanning two years from the earthquake (fitting period) and compare that PSD model with observation of post-fitting period. As a result, we found that their difference were smaller than 1cm for horizontal component, 2cm for vertical component at 2 years after the end of fitting period when Log+Exp+Log model was used.