T233 Landslide Hazard and Risk Assessments II

Thursday, 29 October 2020
5:30 PM - 8:00 PM
Meeting Rooms

EVALUATION OF THE HAZMAPPER GOOGLE EARTH ENGINE APPLICATION FOR COSEISMIC LANDSLIDE MAPPING USING THE MW7.8 2016 KAIKŌURA, NEW ZEALAND EARTHQUAKE AS A CASE STUDY

Paper number: T233-3
Program: Technical Sessions
Time: 5:50 PM – 6:05 PM

Authors

Abstract

Modern satellite networks with rapid repeat-cycles allow for near-real-time imaging of areas impacted by natural hazards such as mass wasting, flooding, and volcanic eruptions. Publicly accessible multi-spectral datasets (e.g., Landsat, Sentinel-2) are particularly helpful in analyzing the spatial extent of disturbances; however, the datasets are large and require intensive processing on high-powered computers, often with the aid of commercial software (e.g., ESRI, ENVI, or ERDAS products). Herein we introduce HazMapper, an open-access multi-hazard mapping application developed and run in Google Earth Engine. HazMapper allows users to derive map and GIS-based products from global Sentinel (10m-pixel) or Landsat (30m-pixel) datasets (1999 to present), requiring only an internet connection, a web browser, and a few minutes of processing time utilizing remote Google servers. HazMapper exploits changes in surface vegetation before and after landslides, or other natural hazard events, to highlight areas of potential impact. The relative difference of NDVI pixel values (rdNDVI) is computed from greenest-pixel mosaic images that bracket the landslide triggering event date by user-defined pre-and-post event time intervals. Additional details at https://hazmapper.org/.

The Mw7.8 14 November 2016 Kaikōura earthquake was the second-largest since the European settlement of New Zealand. A hand-digitized post-earthquake inventory developed from high-resolution satellite and aerial images identified 10,195 landslides triggered by this earthquake across an area of ~10,000 km2 (Massey et al., 2018; DOI:10.1785/0120170305). We use this inventory as the account of record against which the performance of HazMapper is evaluated in its effectiveness for semi-automated landslide inventory generation from a remote, mountainous location. HazMapper performs well for the identification of landslides as small as 100 m2 from lower elevations below the climatically controlled tree line (~1400 m), but struggles in the identification of slides from higher elevations where vegetation is sparse, and snow cover persists. A posteriori knowledge of rdNDVI pixel changes derived from both natural and anthropogenic causes helps decipher HazMapper results for the creation of event-based landslide inventories.

Geological Society of America Abstracts with Programs. Vol. 52, No. 6

doi: 10.1130/abs/2020AM-359135

© Copyright 2020 The Geological Society of America (GSA), all rights reserved. Permission is hereby granted to the author(s) of this abstract to reproduce and distribute it freely, for noncommercial purposes. Permission is hereby granted to any individual scientist to download a single copy of this electronic file and reproduce up to 20 paper copies for noncommercial purposes advancing science and education, including classroom use, providing all reproductions include the complete content shown here, including the author information. All other forms of reproduction and/or transmittal are prohibited without written permission from GSA Copyright Permissions.