面向地震动模型构建的MEMS记录适用性评估方法

A method for assessing the usability of MEMS records for ground-motion model construction

摘要

随着微机电系统(micro-electro-mechanical systems，MEMS)加速度计在预警与密集台网中的广泛部署，其记录能否用于地震动模型(ground-motion model，GMM)构建亟须系统验证。本文面向傅里叶有效幅值谱地震动模型构建，提出一套记录频谱可用性验证与差异归因分析框架。该框架在统一数据处理流程和一致模型形式基础上，通过基准集与扩展集的对照回归识别模型差异，并结合残差结构分析与预测表现比对，区分差异来源于仪器系统偏差还是台站覆盖变化。以四川地区2021-2024年力平衡式加速度计(force-balance accelerometer，FBA)与MEMS加速度记录为例，构建基准数据集(FBA)与扩展数据集(FBA+MEMS)，并采用简化GMM逐频点混合效应回归。研究结果表明，2套模型除常数项与几何扩散项存在一定差异外，其余系数总体一致；扩展集引入后的变化主要体现为台站间残差的调整，而事件间与台站内残差变化有限，且水平竖向谱比法(horizontal-to-vertical spectral ratio，HVSR)统计提示差异更可能源于台站场地覆盖变化而非仪器系统偏差。本文结果为MEMS记录在区域GMM构建中的应用提供了实证依据，所提出的方法可为新增数据的GMM适用性评估提供技术框架。

Abstract

With the widespread deployment of micro-electro-mechanical systems (MEMS) accelerometers in earthquake early warning and dense seismic networks, the usability of their records for the development of ground-motion models (GMMs) requires systematic validation. This study proposes a framework for validating spectral-record usability and attributing model differences for the development of Fourier effective amplitude spectrum ground-motion models (EAS-GMMs). Based on a unified data-processing procedure and a consistent model form, the framework identifies model differences through comparative regressions between a reference dataset and an expanded dataset, and further distinguishes whether such differences arise from instrumental bias or changes in station coverage by combining residual-structure analysis with predictive-performance comparison. Using force-balance accelerometer (FBA) and MEMS acceleration records from Sichuan, China, during 2021-2024 as an example, a reference dataset (FBA) and an expanded dataset (FBA+MEMS) were constructed, and frequency-by-frequency mixed-effects regressions were performed using a simplified GMM. The results show that, except for some differences in the constant term and geometric spreading term, the remaining coefficients are generally consistent between the two models. The changes introduced by the expanded dataset are mainly reflected in adjustments to between-station residuals, whereas changes in between-event and within-station residuals are limited. In addition, horizontal-to-vertical spectral ratio (HVSR) statistics suggest that these differences are more likely associated with changes in site-coverage characteristics of the stations rather than systematic instrumental bias. The results provide empirical support for the use of MEMS records in regional GMM development, and the proposed framework offers a methodological basis for evaluating the applicability of newly introduced data to GMM construction.

关键词

MEMS记录 / 地震动模型 / 有效幅值谱 / 混合效应回归

Key words

MEMS records / ground motion model / effective amplitude spectrum / mixed effect regression

参考文献

[1] Lawrence J F, Cochran E S, Chung A, et al. Rapid earthquake characterization using MEMS accelerometers and volunteer hosts following the M7.2 Darfield, New Zealand, earthquake[J]. Bulletin of the Seismological Society of America, 2014, 104(1): 184-192.

[2] Clayton R W, Heaton T, Kohler M, et al. Community seismic network: a dense array to sense earthquake strong motion[J]. Seismological Research Letters, 2015, 86(5): 1354-1363.

[3] Xie Junju, Li Xiaojun, Wen Zengping, et al. Soil profile database and site classification for national strong-motion stations in western China[J]. Seismological Research Letters, 2022, 93(3): 1930-1942.

[4] Evans J R, Allen R M, Chung A I, et al. Performance of several low-cost accelerometers[J]. Seismological Research Letters, 2014, 85(1): 147-158.

[5] 郭文轩, 任叶飞, 姚鑫鑫, 等 . 简易烈度计记录的噪声水平评价: 以泸定M_S6.8级地震为例 [J]. 振动与冲击, 2024, 43(18): 313-320.

[6] Guo Wenxuan, Ren Yefei, Yao Xinxin, et al. Evaluating on the noise level of MEMS record: case study in Luding M_S6.8 earthquake [J]. Journal of Vibration and Shock, 2024, 43(18): 313-320. (in Chinese)

[7] Wang Wencai, Xie Junju, Li Xiaojun . Usability of records by China earthquake early warning network for ground motion modeling[J]. Soil Dynamics and Earthquake Engineering, 2025, 196: 109497.

[8] Mohammed S, Shams R, Nweke C C, et al. Usability of community seismic network recordings for ground-motion modeling[J]. Earthquake Spectra, 2024, 40(4): 2598-2622.

[9] Bayless J, Abrahamson N A . Summary of the BA18 ground-motion model for Fourier amplitude spectra for crustal earthquakes in California[J]. Bulletin of the Seismological Society of America, 2019, 109(5): 2088-2105.

[10] Bora S S, Cotton F, Scherbaum F . NGA-West2 empirical Fourier and duration models to generate adjustable response spectra[J]. Earthquake Spectra, 2019, 35(1): 61-93.

[11] Vemula S, RaghuKanth S T G, Ponnalagu A . Fourier amplitude spectrum prediction and generation of synthetic ground motion to New Zealand[J]. Acta Geophysica, 2022, 70(1): 39-70.

[12] Davatgari-Tafreshi M, Pezeshk S, Bora S S . Empirical models for Fourier amplitude spectrum of ground-motion calibrated on data from the Iranian plateau[J]. Bulletin of Earthquake Engineering, 2024, 22(8): 3845-3874.

[13] 任叶飞, 刘霜霜, 郭文轩, 等 . 川滇地区地震动傅里叶幅值谱预测模型[J]. 地球物理学报, 2026, 69(6): 2573-2587.

[14] REN Yefei, LIU Shuangshuang, GUO Wenxuan, et al. Ground motion model of Fourier amplitude spectrum in Sichuan-Yunnan region[J]. Chinese Journal of Geophysics, 2026, 69(6): 2573-2587. (in Chinese)

[15] Yang Zhigao, Xu Tairan, Liang Jianhong . Towards fast focal mechanism inversion of shallow crustal earthquakes in the Chinese mainland[J]. Earthquake Research Advances, 2024, 4(2): 100273.

[16] Guo Wenxuan, Ren Yefei, Kishida T, et al. An automated protocol for filter processing of strong-motion records: case study in China[J]. Earthquake Spectra, 2025, 41(3): 2436-2459.

[17] Ancheta T D, Darragh R B, Stewart J P, et al. NGA-west2 database[J]. Earthquake Spectra, 2014, 30(3): 989-1005.

[18] Bayless J, Abrahamson N A . An empirical model for Fourier amplitude spectra using the NGA-West2 database: PEER Report 2018-07[R]. Berkeley, CA: Pacific Earthquake Engineering Research Center, University of California, Berkeley, 2018.

[19] Konno K, Ohmachi T . Ground-motion characteristics estimated from spectral ratio between horizontal and vertical components of microtremor[J]. Bulletin of the Seismological Society of America, 1998, 88(1): 228-241.

[20] Kottke A R, Abrahamson N A, Boore D M, et al. Selection of random vibration theory procedures for the NGA-East project and ground-motion modeling[J]. Earthquake Spectra, 2021, 37(S1): 1420-1439.

[21] Atkinson G M . Ground-motion prediction equations for Hawaii from a referenced empirical approach[J]. Bulletin of the Seismological Society of America, 2010, 100(2): 751-761.

[22] Zhou Jian, Li Xiaojun, Dai Zhijun, et al. Parametrical model for estimating V_S30 from shallow borehole profiles using a database for China [J]. Bulletin of the Seismological Society of America, 2021, 111(3): 1199-1220.

基金资助

国家自然科学基金面上项目(52478567)

黑龙江省自然科学基金杰出青年项目(JQ2023E002)

中国地震局2026年度地震监测预警工作任务项目(CEA-JCYJ-202601030)