During an earthquake, solid ground can loosen into something like quicksand.
Under earthquakes, some types of soil shake liquid, a softening caused by groundwater pressure that becomes an evil twin of the ground shaking. Liquefaction causes major ground deformations that have toppled buildings large and small, as well as crushed pipes beneath them, taking away roads, railways, bridges, and levees.
Such was the case for the 2010 Canterbury earthquake series, the most damaging of which was the 6.2-magnitude Christchurch earthquake. The series of earthquakes, 21 of which were larger than magnitude 5, caused $3 billion in damage to buildings and infrastructure in New Zealand’s South Island, particularly in the city of Christchurch, which would bear the brunt of the deadlier aftershocks a year later.
In 2021, scientists completed a massive collection of earthquake liquefaction data from the three largest Canterbury earthquakes from 2010-2016.
The dataset, which includes more than 15,000 liquefaction histories, has been made public on the NHERI DesignSafe cyber infrastructure. An accompanying data paper was published in March 2021 in the journal Earthquake Spectra.
The authors of the earthquake liquefaction dataset received a 2021 DesignSafe Dataset Award in recognition of the dataset’s diverse contributions to natural hazard research.
The authors are Mertcan Geyin and Brett W. Maurer of the University of Washington; Brendon A. Bradley of the University of Canterbury; Russell A. Green of Virginia Tech; and Sjoerd van Ballegooy of Tonkin + Taylor Ltd.
This particular dataset documents the soil deformation effects of liquefaction on structures in the Canterbury region of New Zealand. Ground reconnaissance and remote sensing captured observations of the occurrence and severity of liquefaction across the region. Cone penetration testing (CPT), which basically pushes a cone into the soil to understand soil resistivity and soil density, was performed, as well as groundwater monitoring.
What is noteworthy is that prior to the Canterbury liquefaction dataset, there existed only about 250 case histories of all other global earthquakes combined.
“This dataset significantly expands the data available for model training and testing by at least a factor of 50, giving the profession a unique opportunity to advance the science of liquefaction prediction,” said co-author Brett Maurer.
The prediction of fluidity is important in reconstruction and in providing the best engineering solutions in the aftermath of an earthquake.
“DesignSafe provides a prominent and visible platform for communicating and disseminating important data,” explains Maurer. “Our datasets are delivered in both Matlab and Python formats, allowing users to work directly with the data (e.g. to build models) without leaving the DesignSafe platform.”
The post-processed data is presented in a single file as a structure array that allows researchers to easily access and analyze a wealth of information relevant to the free-field liquefaction response.
Computer models of scientists who predict liquefaction need to be trained and tested on real data. The case histories show locations where the liquefaction reaction was observed after the earthquake; where ground movements were recorded or can reasonably be approximated, and where in-situ geotechnical tests were performed to characterize the ability to withstand liquefaction.
The data lifecycle begins when an earthquake occurs, recording ground movements in the affected area. Immediately thereafter, ground reconnaissance and remote sensing capture observations of the occurrence and severity of liquefaction across the region. In the months and years that follow, CPTs and groundwater monitoring are performed.
The case histories are then compiled one by one, carefully studying each CPT; the reconnaissance data and satellite images of each CPT site; the ground motion intensities during each earthquake at each CPT location; and groundwater depths at each CPT site at the time of each earthquake.
“Collecting and processing the data took years of effort,” Maurer says. “However, none of this would have been possible without the hundreds of people who have primarily worked to obtain the data (e.g., CPT testing, satellite imagery, groundwater modeling, etc.) as part of a massive effort funded by the New Zealand Government In relation to that effort, our work to collect and publish the data has been trivial.”
Ultimately, everyone in society benefits from improved hazard assessments.
“In many earthquakes, the liquefaction of the soil causes enormous damage and loss, as evidenced by the earthquakes in Canterbury, New Zealand, which damaged large parts of the city beyond repair and turned green,” Maurer said. “When someone builds a road, bridge, house, etc., building codes require that the liquefaction hazard be assessed. We all want those assessments to be accurate.”
“Data is everything in life. This is how we make decisions about every move we make. Without earthquake data, we don’t know if our current risk prediction models are working and we have no way to make better predictions going forward,” concludes Maurer.
Funding for the dataset was provided by the National Science Foundation (NSF), US Geological Survey (USGS), and Pacific Earthquake Engineering Research Center (PEER) under Grants CMMI-1751216, CMMI-1825189, CMMI-1937984, G18AP-00006, and 1132 -NCTRBM, respectively. The work was also supported by QuakeCoRE, the New Zealand Center for Earthquake Resilience. Above all, the authors acknowledge the countless people who contributed to the collection of data on the history of liquefaction, which was collected under the auspices of the New Zealand Earthquake Commission (EQC).
DesignSafe is a comprehensive cyber infrastructure that is part of the NSF-funded Natural Hazard Engineering Research Infrastructure (NHERI) and provides cloud-based tools to manage, analyze, understand and publish critical data for research to understand the impact of natural hazards. The capabilities within the DesignSafe infrastructure are available free of charge to all researchers working in natural hazards. The cyber infrastructure and software development team is based at the Texas Advanced Computing Center (TACC) at the University of Texas at Austin, with a team of natural hazards researchers from the University of Texas, Florida Institute of Technology and Rice University, comprising the senior management team.
NHERI is supported by multiple grants from the National Science Foundation, including the DesignSafe Cyberinfrastructure, Award #2022469.