During an earthquake, soil can weaken through subsidence and liquefaction. These processes can cause buildings to collapse as the soil becomes unable to support their weight. Researchers have now developed a model that predicts soil-bearing strength and thickness to identify stable construction sites and reduce structural risks. Additionally, the model can also predict other soil conditions in real-time and function as an early-warning system to identify potential hazards.
On September 1, 2024, Japan will observe the 64th Disaster Prevention Day, commemorating the Great Kanto Earthquake which occurred on September 1, 1923. This earthquake, which struck the Kanto region, including the capital Tokyo, caused widespread destruction and claimed the lives of more than 140,000 people. It was a pivotal moment in Japan's history, prompting the nation to prioritize disaster preparedness. These efforts have resulted in Japan becoming the world leader in disaster preparedness, early warning systems, and earthquake-resilient structures.
Buildings on these efforts, Professor Shinya Inazumi and Ph.D. student Yuxin Cong from the Graduate School of Engineering and Science, both at Shibaura Institute of Technology, have developed a system to address risks caused by liquefaction and subsidence, two soil deformation events triggered by earthquakes that cause buildings to tilt and sink. In a study published in the journal Smart Cities on 6 May 2024, they introduce a predictive model that maps the distribution of soil-bearing layers. The model can identify regions suitable for construction, mitigating the risk of structural failure during earthquakes.
"In the broader context of smart cities, the implications of this study are many. Using advanced predictive models, urban planners and engineers can more accurately assess the suitability of sites for development and optimize the design and placement of buildings, infrastructure, and public utilities," explains Prof. Inazumi.
Predicting which regions are vulnerable to these effects is challenging as it is impractical to evaluate soil conditions at every location. To address this, researchers developed a model to predict soil bearing strength and the thickness of bearing layers using geotechnical data from 433 sites in Setagaya, Tokyo. This data was obtained using the Standard Penetration Test and the Mini-Ram Sounding Test, two methods widely used in Japan to assess soil density and foundation requirements.
Using this data, researchers applied the kriging method, a well-known statistical technique, to make predictions of soil-bearing layer thickness and depth based on geographic coordinates such as latitude and longitude. This produced a three-dimensional map showing the distribution of bearing layers at different locations in Setagaya, Tokyo. To further improve prediction accuracy, they used the bagging technique, an ensemble learning method, which combines predictions from multiple models to produce a more accurate result. For this method, they included geographical data such as latitude, longitude, and elevation along with the geotechnical data to improve the predictive capabilities of the model.
The bearing strength and thickness of soil layers are indicators of the soil's ability to support buildings and other heavy structures. A generated map of soil-bearing properties can help city planners ensure structures are built on stable foundations, minimizing the risk of failure during soil deformation events. Additionally, the model can be integrated with real-time data from sensors that monitor parameters such as moisture and ground movement. This allows city planners and engineers to continuously monitor changes in soil conditions and identify potential risks, such as soil instability, which could compromise safety.
Such a system advances disaster risk reduction efforts and contributes to Goal 11 of the UN's Sustainable Development Goals, which seeks to make cities and human settlements more inclusive, safe, resilient, and sustainable. Its implementation is particularly vital in planning resilient cities, especially in earthquake-prone regions like Japan.