PEER Report 2024/06: "Ground Failure of Hydraulic Fills in Chiba, Japan and Data Archival in Community Database"

July 9, 2024

PEER has published Report No. 2024/06: "Ground Failure of Hydraulic Fills in Chiba, Japan and Data Archival in Community Database." It was authored by Scott J. Brandenberg, Jonathan P. Stewart, Kenneth S. Hudson, Dong Youp Kwak, Paolo Zimmaro and Quin Parker, Department of Civil and Environmental Engineering, University of California, Los Angeles, California. 

Access the report and 2 page summary. Access the full listing of PEER reports.

Abstract

This report describes analysis of ground failure and lack thereof observed in the Mihama Ward portion of Chiba, Japan following the 2011 M9.0 Tohoku Earthquake. In conjunction with this work, we have also significantly expanded the laboratory component of the Next Generation Liquefaction (NGL) relational database. The district referred to as Mihama Ward is on ground composed of hydraulic fill sluiced in by pipes, thereby resulting in a gradient of soil coarseness, with coarser soils deposited near the pipes and fine-grained soils carried further away. Observations from local researchers at Chiba University following the 2011 Tohoku Earthquake indicate that ground failure was observed closer to the locations where the pipes deposited the soil, and not further away. This ground failure consisted of extensive sand boiling and ground cracking, which led to building settlement and pipe breaks. Our hypothesis at the outset of the project was that liquefaction susceptibility might explain the pattern of ground failure. Specifically, soils deposited near the pipes are susceptible due to their coarser texture, while soils further from the pipes may be non-susceptible due to the presence of clay minerals and higher plasticity. Were this hypothesis borne out by evidence, soil in the transition zone would have provided important insights about liquefaction susceptibility. Based on testing of soils in our laboratory, we find this hypothesis to be only partially correct. We have confirmed that there are regions with high clay contents and no ground failure and other regions with predominantly granular soils and extensive surface manifestation of liquefaction. Where the hypothesis breaks down is in the transition zone, where we found that the fine-grained soils are non-plastic, and therefore they are susceptible to liquefaction. Our interpretation is that these silt materials likely liquefied during the earthquake, but did not manifest liquefaction. Two factors may have contributed to this lack of manifestation: (1) level ground conditions and lack of large driving static shear stresses (structures in the region are light residential construction) and (2) the silt is less likely to erode to the surface and form silt boils than the sandier soils that produced surface manifestations. This case history points to the importance of separating triggering (defined as the development of significant excess pore pressure and loss of strength) from manifestation (defined as observations of ground failure, including cracking, sand boils, and lateral spreading). The Mihama Ward case history involved laboratory tests performed by Tokyo Soil Research Co. Ltd. and the UCLA geotechnical laboratory. Given the importance of this data to the understanding of this case history, we recognized a need to incorporate laboratory tests in the NGL database alongside field tests and liquefaction observations. We therefore developed an organizational structure for laboratory tests, including direct simple shear, triaxial compression, and consolidation, and implemented the schema in the NGL database. We then uploaded data from tests performed by Tokyo Soil and UCLA. Furthermore, numerous other researchers have also uploaded laboratory test data for other sites. This report describes the organizational structure of the laboratory component of the database, and a tool for interacting with laboratory data