今週のうなぎセミナーについてお知らせいたします。
後期のうなぎゼミが始まります。初回はガイダンスと野末さんの発表です。仮スケジュールですが、ガイダンス時に調整します。

Here is information of the Unagi-seminar(October, 2nd).
We will begin by briefly discussing the dynamics that you are all familiar with and the schedule of presentations. We will also have the first presentation of the seminar program.

************** Seminar on Seismology IV B, D /地震学ゼミナールIV B, D (Unagi Seminar) **************

科目：地震学ゼミナールIV B, D / Seminar on Seismology IV B, D（修士・博士）
日時：2025年 10月 2日 (木) 13:30~
場所：京都大学 防災研究所 本館E-232D

Date and Time：2025-10-02(Thursday), 13:30~
Place：Uji Campus Main Building E232D

---------- ---------- ---------- ----------

Speaker（発表者）:　野末 陽平（Yohei NOZUE）
Title（題目）:
Estimation of a strain-rate field in southeastern Tibet by joint inversion of GNSS, leveling, and InSAR data using basis function expansion

Abstract（要旨）:
　Strain accumulation is related to earthquake occurrence. Therefore, it is important to estimate crustal deformation from geodetic data. By several methods including the basis function expansion, strain-rate fields have been estimated from GNSS data in Japan (e.g., Sagiya et al. 2000; Okazaki et al. 2021; Nozue & Fukahata, 2025). However, in foreign countries, GNSS observations are not necessarily distributed with enough density to monitor crustal deformation in detail. On the other hand, InSAR observes the line-of-sight (LOS) displacements with very high spatial resolution. Therefore, it is expected that combining them will contribute to higher resolution and accuracy in estimating deformation fields. In this study, we focus on southeastern Tibet, which is an active tectonic zone and has numerous faults, and jointly invert GNSS, leveling, and InSAR data to estimate strain-rate fields. 　We use the average velocities from 2014 to 2023. GNSS 3D or 2D velocities and leveling changes are summarized in Elliot et al. (2025). We use InSAR LOS velocities of the four ascending and four descending frames provided by the COMET in the UK. 　Following Yabuki & Matsu’ura (1992) and Okazaki et al. (2021), we estimate a strain-rate field by using the method of basis function expansion. The model velocity field is expressed by the superposition of the basis functions and connected to the observed velocity data in the objective function. Since leveling observes displacements relative to a reference point, the correction to the absolute level change is needed. InSAR observes the LOS displacements, which requires the projection of the basis functions to the LOS direction. In addition, InSAR data is also influenced by the orbital errors, which are usually expressed by a quadratic polynomial function of coordinates. Therefore, we incorporate the matrix for these projections and corrections into the observation equation, and then estimate the correction parameters, as well as model parameters that express crustal deformation, in the following analysis. We define the objective function, where the first term is composed of the residual sum of squares between observed and calculated velocities, and the second term is the prior constraints on the smoothness of the velocity fields. The relative weight between the two terms is regulated by a hyperparameter, whose value is optimized by ABIC minimum (Akaike, 1980). By minimizing the objective function, we estimate the model parameters and correction parameters for leveling and InSAR. By analytically differentiating the velocity field, we obtain the strain-rate field. 　The estimated strain-rate field shows the strain localization along the Xianshuihe fault, which was alsp found by previous studies (e.g., Fang et al. 2024). The peak value of maximum shear strain rates is ~150 nanostrain/yr, which is higher than that estimated from only GNSS data. 　I worked on this study during the visit to the University of Leeds in the UK from April to August. I will also briefly talk about life in Leeds.