The application of remotely sensed soil moisture for landslide hazard early warnings
Abstract
Landslide is one of the most common and dangerous natural hazards worldwide. Among most types of landslides, soil moisture plays a vital role in their triggering. However in-situ soil moisture data are rarely available in hazardous zones. The advanced remote sensing technology could provide useful soil moisture information. In this study, an assessment has been carried out between the latest version of the European Space Agency Climate Change Initiative soil moisture product (CCI-SM) and the landslides events in a northern Italian region in the 14-year period 2002-2015. The results showed a good correlation between the CCI-SM and the in-situ observations (r and RMSE are calculated as 0.68 and 0.06 m3/m3 respectively). An evident correlation has been found between the CCI-SM and the landslides events, as over four-fifths of events had soil wetness conditions above the 50% regional soil moisture line. Attempts have also been made to explore the soil moisture thresholds for landslides occurrences under different environmental conditions (i.e., land cover, soil type, and slope). The results showed slope distribution could provide a rather distinct separation of the soil moisture thresholds, with thresholds becoming smaller for steeper areas, indicating dryer soil condition could trigger landslides at hilly areas than in plain areas. The thresholds validation procedure is then carried out. 45 rainfall events between 2014-2015 are used as test cases. Contingency tables, statistical indicators, and Receiver Operating Characteristic (ROC) analysis for thresholds under different exceedance probabilities (1% - 50%) are explored. The results have shown that the thresholds using 30% exceedance probability provide the best performance with the hitting rate at 0.92 and false alarm at 0.50. We expect this study can provide useful information for adopting remotely sensed soil moisture in landslides early warnings.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2018
- Bibcode:
- 2018AGUFMNH43D1085H
- Keywords:
-
- 4306 Multihazards;
- NATURAL HAZARDSDE: 4333 Disaster risk analysis and assessment;
- NATURAL HAZARDSDE: 4336 Economic impacts of disasters;
- NATURAL HAZARDSDE: 4352 Interaction between science and disaster management authorities;
- NATURAL HAZARDS