Integrated Earthquake Hazard Assessment System with Geotechnical Spatial Grid Information Based on GIS

Abstract

최근 국내외적으로 지진발생 빈도와 규모가 증가하고 있으며, 이에 따른 지진과 지진재해에 관한 관심과 우려가 증대되고 있다. 이에 따라 구가기반시설을 대상으로 지진발생 및 예상피해 규모를 사전에 예측할 수 있는 시스템의 개발이 시도되고 있다. 그러나 지반과 구조물의 내진 성능을 체계적으로 평가하고, 지진발생 시에 지진계측 정보를 연계함으로써, 지진재해를 실시간으로 예측하는 기술 개발은 미비하다. 본 연구에서는 지진발생 시에 지반공학적 지진재해를 준실시간으로 평가할 수 있는 체계를 수립하고, 이를 바탕으로 GIS기반의 지반공학적 공간그리드 정보를 활용한 지진재해평가 통합정보화 시스템을 개발하였다. 평가 시스템은 데이터베이스(DB)를 중심으로 현장 자료를 표준화하여 DB화할 수 있는 입력 모듈과, 신뢰도가 확보된 3차원의 지반정보를 결정할 수 있는 지구통계학적 3차원 통합분석 모듈, 부지응답 특성을 고려함으로써 액상화와 구조물 취약도를 평가할 수 있는 실시간 지진재해평가 모듈, 그리고 3차원 출력 및 가시화 모듈로 구성된다. 먼저 입력모듈을 통해 지형, 지반, 구조물 정보와, 실시간의 지진계측 정보를 공간정보와 연계하여 DB(geodatabase)에 입력하며, 구축 DB는 국가지진관측망 서버에 탑재되어 운용 중에 있다. 지구통계학적 3차원 통합분석 방법 및 모듈에 따라, 지반정보 중 시주조사 자료의 신뢰도 확보를 위해 이상치 검증기법(교차검증, 극한분포)을 개발하였으며, 이를 이용하여 시추조사 자료를 최적화한다. 최적화된 시추조사 자료와 함께 인접 영역에서 확보한 물리탐사 토모그래피를 통합 활용할 수 있도록 지시자 크리깅 기법을 통한 3차원 통합분석 방법을 개발하였다. 끝으로 지진재해평가 방법과의 연계를 위해 3차원 지반정보를 세분화하여 지반동적 특성값을 부여함으로써 지반공학적 공간그리드를 구축한다. 구축된 3차원의 지반공학적 공간그리드를 바탕으로 준실시간으로 지진재해를 평가할 수 있는 방법 및 모듈을 개발하였다. 대상영역의 부지응답 특성을 고려하기 위해, 계측된 암반가속도를 입력함으로써, 실시간으로 지반의 최대가속도를 결정할 수 있는 방법을 개발하였다. 이후 액상화 가능지수(LPI)를 이용하여 실시간으로 액상화 피해 정도의 공간분포를 평가할 수 있는 방법과 구조물의 취약도 함수와의 연계를 통해 구조물 손상도를 실시간으로 평가할 수 있는 방법을 개발하였다. 수립된 지반공학적 지진재해평가 체계에 따라 지진발생 시에 자동화된 일련의 지진계측 정보 연계 및 신속한 재해평가가 가능하도록 GIS기반의 통합정보화 시스템 프로그램을 개발하였다. 개발시스템의 적용성 검증 차원에서 국내 주요 항만지역(인천항, 부산항, 부산신항)을 대상으로 지진재해 평가 시뮬레이션을 수행하였으며, 즉각적인 재해결과를 예보할 수 있었다. 본 개발 시스템은 국가지진관측망에 탑재되어, 향후 지진발생 시에 지진재해를 선재적이고 즉각적으로 예보함으로써 합리적인 대응 및 복구방안 수립의 기초자료로써 활용되고 있다.

Historically, the Korean peninsula has been regarded as a safe region with respect to the hazard of earthquakes due to the characteristics of its seismotectonic location, being classified as a region of moderate seismicity. However, the number of earthquake events keeps increasing every year, and the recent cases of earthquake hazards invoke the necessity of seismic study in Korea, as geotechnical earthquake hazards, such as strong ground motion, liquefaction and landslides, are a significant threat to structures in port or downtown areas built on seismically vulnerable loose and saturated sandy soils. Therefore, a seismic disaster management system (platform) is required to establish effective and appropriate strategies to reduce earthquake hazards. In this study, an integrated earthquake hazard assessment system with geotechnical spatial grid information was developed based on a geographic information system (GIS). The developed system built, within the frame of GIS, consists of a database (DB) containing all site information and processed data in the system in the standard data formats, and the system software performing various functions to manage and utilize the data in the database. The system software is functionally divided into an input module, a geostatistical three-dimensional (3D) integration module, a real-time earthquake hazard assessment module, and output or visualization module. The database and these modules of the system software were combined and integrated into a single system to provide a familiar and user-friendly working environment with a standard interface. In addition, the integrated system was imbedded into a Korea Integrated Seismic System (KISS) server to be linked with real-time seismic accelerations. DB is the backbone of the developed framework. It stores not only primary data such as geography, geotechnical investigation results, man-made structures, and seismic monitoring data which are transmitted in real time from a seismometer server (at KISS), but also secondary data obtained from a geostatistical 3D integration module and real-time earthquake hazard assessment module. It contains alphanumeric datasets according the outcome of data classification and standardization with a geodatabase (GDB). The data stored in the database can be easily utilized in this framework. Input function provides an effective way to store and organize all collected field data, including electric or non-electric documents, geographic and geotechnical investigations, man-made structures, seismic monitoring, and analysis data, according to a standard data format based on GDB. Geotechnical spatial grids in the 3D domain are constructed for target sites with geostatistical methods in a process known as geostatistical 3D integration. This method has three functional modules with the database: an outlier detection module, a geostatistical integration module, and geotechnical spatial grid construction module. Geotechnical investigation results always reflect the level of soil uncertainty. To handle this uncertainty, the outlier detection methods, which optimize the borehole datasets, are conducted. The geostatistical integration method based on indicator kriging is performed using optimized borehole datasets and digitized geophysical tomography results to construct 3D geo-layers. The 3D geo-layers are categorized and subdivided into representative soil profiles with dynamic properties to add 3D geotechnical spatial grid to the database. This step must be conducted as a baseline prior to the occurrence of earthquakes. The real-time framework for integrated earthquake hazard assessment with geotechnical spatial grid information has three functional modules with the database: a real-time seismic load determination module, a real-time liquefaction hazard estimation module, and a real-time structure fragility evaluation module. In the first, linked with the 3D geotechnical spatial grid, correlations between rock outcrop accelerations and the maximum accelerations of each layer considering the site response characteristics are predetermined. Thus, as earthquake events occur, as soon as monitored rock outcrop acceleration data are transmitted from the accelerometer, the seismic load at each spatial grid is estimated. In the second, the potential damage due to liquefaction is estimated by combining the geotechnical spatial grid and correlated maximum acceleration of each layer based on the simplified liquefaction potential index (LPI) evaluation method in real time. Finally, in the third, the possibility of structure failure is evaluated by the integration of the geotechnical spatial grid and correlated maximum acceleration based on the structure fragility curve in real time. The output and visualization module displays all attributive information in the database using tables and graphics according to its characteristics. Also, all data in the database can be output as a chart or a graphic. The graphic functions, such as 2D plane view, 2D sectional view, and 3D view display, interpolated data are shown in several display formats at same time. Specifically, the earthquake hazard assessments results are visualized to 2D or 3D GIS-maps with satellite images, and the seismic damage (composed of the seismic load, liquefaction potential, and structural fragility) of the target structure can be calculated using zonation criteria in real time. A simulation of the developed system was specifically conducted at Busan port, Korea, using two virtual earthquake scenarios (Uljin and Tohoku Earthquakes) and two actual earthquake events (2013 Baengnyeong and 2014 Taean Earthquakes) of the Incheon port. The simulation results are visualized as a geotechnical earthquake hazard map to verify the computer-aided real-time assessment framework. A verification test of the integrated system was also performed for the Incheon port using data actually transmitted from the accelerometer of the KISS server when two notable earthquake events occurred at the nearby Incheon port.