Using network science to evaluate vulnerability of landslides on Big Sur Coast, California, USA

Landslide events, ranging from slips to catastrophic failures, pose significant challenges for prediction. In this study, a physically inspired framework is employed to assess landslide vulnerability at a regional scale (Big Sur Coast, California). Our approach integrates techniques from the study of complex systems combined with multivariate statistical analysis to identify unstable areas vulnerable to landslide events. We successfully apply a technique originally developed on the 2017 Mud Creek landslide, Big Sur, and refine our statistical metrics to characterize landslide vulnerability within a larger geographical area. Our results successfully classify four landslide events that occurred in the winter year of 2022-2023 as areas that are vulnerable to slope failure. The performance of our methods is compared to factors such as landslide location, slope, cumulative displacement, precipitation, and InSAR coherence, via a multivariate statistical analysis. We conclude that our network ..., Open-source code R is used to create the networks and Matlab to run the community detection algorithm. The code can be found on https://github.com/vddesai-97/networkLandslide.git, which uses the community detection algorithim from https://github.com/GenLouvain/GenLouvain. The dataset contains 17 sub-regions with corresponding edge lists, spatial grids, and edge weights. In addition, the interferogram list for the InSAR data used for analysis and the shapefiles for the 44 landslide polygons are included., , # Data from: Using network science to evaluate vulnerability of landslides on Big Sur Coast, California, USA

[https://doi.org/10.5061/dryad.1jwstqk42](https://doi.org/10.5061/dryad.1jwstqk42)

This folder contains edge lists, spatial coordinates, shapefiles, and multilayer weighted edges for the 17 sub-regions used in the paper.

There are two components to the data -- the spatial coordinates of each node, which represents a patch of area, and then the edge list which identifies node pairs that are connected based on nearest neighbors.

Information about the system, such as velocity and slope, is then connected to the edges as weights, which is stored as an matrix where each row represents an edge, the first two columns represents the node pairs, and then the preceding columns represent information for each time layer within the dataset corresponding to the interferogram list.

There are 284 InSAR time slices, and the corresponding combinations of SAR images used are included in the in...