Morgenstern, R.; Rosser, B.J.; Lukovic, B.; Zoeller, A.; Scadden, P.G.; Cox, S.C. 2025 The New Zealand Landslide Database version 2: a data dictionary. Lower Hutt, N.Z.: Earth Sciences New Zealand. GNS Science report 2024/49. 42 p.; doi: 10.21420/4Z6H-1F97
Abstract
fundamental for developing landslide hazard and risk models at all scales. To better understand the factors that trigger landslides, as well as their movement mechanisms and physical properties, it is important to capture and document where landslides have occurred in the past so that their impacts can be better mitigated in the future. To achieve this, the New Zealand Landslide Database (NZLD)– a national geospatial inventory containing landslide-related data – was previously established to provide factual, standardised data that can be used reliably for landslide hazard and risk assessments. Since its release, a number of issues were identified with the database that impact its useability. These issues included that there was an absence of some key information, it was difficult to load bulk datasets and edit existing data, it was impossible to relate landslide features together and there was an absence of quality assurance information and some metadata. The aim of this project was to resolve these and other minor issues, to upgrade the NZLD (to version 2) to make it more fit-for-purpose andto create a Geographic Information System (GIS) template compatible with the NZLD that can be used systematically for future landslide data collection (i.e. landslide mapping). The database structure has been refined and the new relational database contains a number of spatial and attribute tables that allow different landslide features (entities) to be related, queried, extracted and utilised for landslide hazard and risk modelling. Landslide features are presented with consistent attributes – described in this data dictionary – to standardise landslide information being stored and collected at a national scale. Landslide-generating events (e.g. earthquakes) are now able to be recorded and matched to landslide entities. Metadata are recorded in a standardised way, and a quality ranking scheme has been developed to provide an overview of the quality of each source dataset. Curators can easily edit the database through ArcGIS, and a new stand-alone ‘Load Tool’ application streamlines the bulk upload process of large datasets and provides automatic quality-control checks. An upgraded webmap application allows the database to be visualised, queried and extracted via a user-friendly interface. (auths)
