According to a group of researchers from the University of California, satellite data give crucial and reliable information for identifying, especially in remote mountainous regions, hotspots for landslides as well as for predicting these events, reducing therefore their potentially devastating effects.
Recognizing the limitations of ground-based observations in many developing countries due to lack of investment, researchers created a model that makes use of satellite data, and therefore a global reach that includes remote and topographically complex regions. "Landslides typically occur in mountainous regions where other sources of information, including radar and gauge measurements (used in standard global landslide models), are not available," Amir AghaKouchak, co-author and assistant professor at the Center for Hydrometeorology and Remote Sensing in Irvine, tells in a press release.
Indeed, the model uses satellite data on rainfall, topographical features of slopes, and land cover. Once tested on a dataset of previous landslides it will help predicting landslides, constituting the basis of a real-time, global landslide prediction system. However, the model also presents some limitations: it "cannot be considered as a general landslide model", since it does not cover earthquake-triggered landslides nor small-scale landslides either; furthermore, dense vegetation may for many areas represent a limitation and add uncertainty to the method. Nevertheless, the model can definitely help to improve landslide monitoring and preparedness.
