Shillong, Feb 4: Researchers at North-Eastern Hill University’s Department of Information Technology have created an AI-based Landslide Susceptibility Map (LSM) for Meghalaya, leveraging an ensemble Machine Learning framework combining ten models.
This innovative approach has yielded an impressive accuracy of over 90% in predicting landslide-prone zones.
“Identifying and regularly monitoring vulnerable areas can significantly reduce the impact of landslides,” said experts, highlighting the importance of this research.
The LSM categorizes Meghalaya’s landslide susceptibility into five risk categories, with approximately 7% of the state falling under the very high-risk category.
East Khasi Hills district is the most vulnerable region, with around 730 km² under very high risk. Other districts at risk include Ri Bhoi, Eastern West Khasi Hills, and West Jaintia Hills. Proximity to roads is the leading cause of landslides, attributed to slope destabilization during construction and altered drainage patterns.
“This research marks a significant advancement in improving public safety and reducing landslide-related hazards in Meghalaya,” said Dr. K. Amitab, lead researcher.
The LSM will aid disaster management agencies in prioritizing resource allocation and guiding proactive planning.
