A spatiotemporal object-oriented data model for landslides (LOOM)

LOOM (landslide object-oriented model) is here presented as a data structure for landslide inventories based on the object-oriented paradigm. It aims at the effective storage, in a single dataset, of the complex spatial and temporal relations between landslides recorded and mapped in an area and at their manipulation. Spatial relations are handled through a hierarchical classification based on topological rules and two levels of aggregation are defined: (i) landslide complexes, grouping spatially connected landslides of the same type, and (ii) landslide systems, merging landslides of any type sharing a spatial connection. For the aggregation procedure, a minimal functional interaction between landslide objects has been defined as a spatial overlap between objects. Temporal characterization of landslides is achieved by assigning to each object an exact date or a time range for its occurrence, integrating both the time frame and the event-based approaches. The sum of spatial integrity and temporal characterization ensures the storage of vertical relations between landslides, so that the superimposition of events can be easily retrieved querying the temporal dataset. The here proposed methodology for landslides inventorying has been tested on selected case studies in the Cilento UNESCO Global Geopark (Italy). We demonstrate that the proposed LOOM model avoids data fragmentation or redundancy and topological inconsistency between the digital data and the real-world features. This application revealed to be powerful for the reconstruction of the gravity-induced deformation history of hillslopes, thus for the prediction of their evolution.

Automated summary

LOOM (landslide object-oriented model) is proposed as a data structure for landslide inventories, aiming to effectively store and manipulate complex spatial and temporal relations between landslides. It uses hierarchical classification and aggregation levels to handle spatial relations, while temporal characterization is achieved by assigning exact dates or time ranges to each object. Tested in the Cilento UNESCO Global Geopark (Italy), the model proves powerful for inventorying landslides and reconstructing gravity-induced deformation history, leading to prediction of evolution.