Software for quantifying shallow landslide hazards to transportation infrastructure under changing climate and forest management

Critical infrastructure in upland landscapes can be damaged by landslides and debris flows, and managing and mitigating the risks posed to this infrastructure is a key challenge for infrastructure owners. In addition, landslides and debris flows do not respect ownership boundaries, and third party landowners can find themselves liable for mass movements originating from their land, whereas infrastructure owners have to manage hazards from beyond their holdings. In addition, transport owners and maintaining agents must be able to plan on works to be delivered at the right time and in the right place. Planning efficient spending has a number of challenges, including persistent uncertainty in how changes in climate and land use affect future landslide susceptibility and limited capability to make effective use of high resolution digital datasets (e.g. LiDAR surveys) to model landslide scenarios.
This project will develop software for producing landslide hazard maps at regional scales incorporating the highest resolution topographic and land use data. It builds on previous hazard research at the University of Edinburgh and the British Geological Survey, and is built to work in a mulitcore, supercomputing environment--the software can handle big data in the form of very high resolution LiDAR surveys that provide information on slopes at the sub-metre scale: the scale at which mass movements and the hydrology that drive them operate. In addition, the software will be designed to provide scenario modelling. In particular, one of the major drivers in changes in landslide hazard is changing root cohesion from different vegetation patterns along transport corridors. The project will assimilate forestry data from Forest Research so that we can investigate the planting and harvesting schedule that minimised landslide risk, and also determine the time at which transport corridors are most threatened by landsliding. The ultimate objective of the project is to provide a software tool that land owners, infrastructure owners and maintaining agents can use to better plan mitigation and spending strategies.

The project will design software that enables end-users to model slope instability along specified transport corridors and evaluate the magnitude of changing landslide susceptibility in response to scenarios of land use and climate change. The software will ingest topographic (e.g., BGS Nextmap, EA LiDAR) and land use data (e.g. secondary roads, vegetation type) and run state-of-the-art slope stability analysis. The outputs are hazard data layers in standard GIS formats. This software will form a step-change in performance and scenario modelling opportunities by enabling high-resolution, quick slope stability modelling using more detailed representations of topography and land use/climate change than is currently available. The software is designed to be used on multicore supercomputers available at Edinburgh and BGS. It does not depend on intermediary GIS software as this is unable to efficiently cope with big data. Automation of scenario modelling will enable combined assessment of land use changes (e.g. cutting schedules of Forestry Commissions) and projected climate (e.g. derived from UKCP09) to determine changes in landslide magnitude and frequency along infrastructure corridors. It will therefore form a unique tool for asset owners who need to manage whole life costs of their assets and third partly land owners (e.g., the Forestry Commission) who need to better understand their long-term liabilities.