Science & Technology, Australia (Commonwealth Union) – Landslides are a common and destructive occurrence taking place around the world. They are defined as the movement of a mass of rock, debris, or earth down a slope. This geological hazard can be triggered by various factors, including heavy rainfall, earthquakes, volcanic eruptions, and human activities. The consequences of landslides can be devastating, resulting in loss of life, destruction of property, and disruption of infrastructure.
A combination of hazard mapping, engineering solutions are generally applied to track landslides, but with the recent development of Artificial Intelligence this tracking process has been enhanced.
Researchers from the University of Melbourne are aiding villagers in Nepal by predicting imminent landslides that threaten their homes.
Scientists from the University of Melbourne, Tribhuvan University in Nepal, as well as the University of Florence have partnered with the Government of Nepal and Australia’s Department of Foreign Affairs and Trade to develop a cutting-edge artificial intelligence system. This system analyzes extensive data to determine when rain-saturated ground is likely to collapse.
The forecasting technology that is known as SAFE-RISCCS, continuously processes satellite images from NASA, the European Space Agency, and the Japan Aerospace Exploration Agency. By integrating SAFE-RISCCS forecasts into Landslide Early Warning Systems, the accuracy of alerts is significantly enhanced, providing warnings days or even weeks ahead of a possible occurrence.
Leading the project, University of Melbourne scientist Professor Antoinette Tordesillas highlighted that approximately 59 percent of Nepal is susceptible to landslides, resulting in one of the highest per capita death rates due to landslides globally.
According to Professor Tordesillas more than 80 percent of Nepal’s land is inclined, and the 2015 Gorkha earthquake significantly destabilized much of the country. As monsoons are expected in June, they are providing assistance to policymakers and risk managers in preparing for future monsoons, which are anticipated to cause more catastrophic landslides due to increasingly frequent and heavy rainfall.
She further indicated that climate change and increasing human activities pose a substantial threat to communities in Nepal, with women, children, the elderly, and the disabled being particularly vulnerable during disasters. Early warning systems are vital as they provide public officials with more time to plan, prepare, and protect communities before the hazard strikes.
“It’s like watching the water wash away sandcastles at the beach; one minute a village is there and the next it is gone. Ground motion holds clues to when a slope is about to slide. Our job is to help develop reliable early warning systems through better forecasting tools that can read these clues and predict landslide hazards as early as possible.”
The SAFE-RISCCS platform leverages satellite imagery and a novel open-access AI tool developed at the University of Melbourne. This technology integrates rainfall measurements with dynamic ground motion data to continuously monitor and predict landslide risks at specific locations and times.
Researchers of the study indicated that SAFE-RISCCS will be deployed in two high-risk areas in collaboration with Nepal’s National Disaster Risk Reduction and Management Authority, as part of a new Landslide Early Warning System being produced in Nepal.
On April 25, 2015, a 7.8 magnitude earthquake struck near Kathmandu, Nepal, triggering avalanches on Mt. Everest and in the Langtang Valley, along with approximately 45,000 landslides that devastated parts of the city and obliterated thousands of villages.
Professor Basanta Adhikari from Tribhuvan University emphasized that continuous monitoring of surface deformations is crucial for protecting the people of Nepal.
Professor Adhikari indicated that the escalating frequency of landslides in the Nepal Himalayas, fueled by the combined impact of seismotectonic activities and climate change, in addition to human-induced factors such as disorganized road construction, underscores the susceptibility of mountain communities.
Professor Emanuele Intrieri of Florence University asserted that fostering awareness is the most efficient means of mitigating risk.
Professor Intrieri further indicated that advancements in remote sensing technology permit them to monitor ground movements from satellites with escalating precision. Enhancing public awareness concerning the risks in their immediate vicinity continues to be the most economical solution to decrease vulnerability.
“Nepal is only one of several regions in the Asia-Pacific at risk of worsening natural disasters associated with climate change. By improving early landslide prediction globally, we reduce the risk of catastrophic events and so protect human life, people’s livelihoods, critical infrastructure and the environment,” added Professor Tordesillas.
