Machine learning can help us predict landslides caused by climate change

Christoph Mertz has a position as a project scientist at Carnegie Mellon University, and he spent months gathering photographs of various hillsides. These photos showed that these areas are becoming wetter and wetter as time goes on. Mertz was determined to develop a machine that would detect signs of landslides becoming active such as cracks in the ground or tilting trees. The current model that him and his team have developed allow these risk factors to be identified for necessary policy adjustments and budget allocation.

Key Takeaways:

Although landslides are a natural occurrence, favorable conditions for their occurrence can be manmade.
When human activity alters a natural slope, or reroutes where waterways, or where rain runoff generally goes, the conditions for a landslide may be initiated.
In 2018, the city of Pittsburgh used up the one million in funds allocated to deal with landslide damage in a matter of months.

“Combined with increased rainfall rates related to climate change, landslides in the United States have become more common and more severe.”