Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Skip to content

Earthquake-Triggered Ground Failure associated with the M7.1 2018 Southcentral, Alaska Earthquake (ver. 2.0, December 2023)

Metadata Updated: December 31, 2023

The November 30, 2018, magnitude (Mw) 7.1 Anchorage, Alaska earthquake triggered substantial ground failure throughout Anchorage and surrounding areas (Grant and others, 2020; Jibson and others, 2020). The earthquake was an intraslab event with a focal depth of about 47 km and an epicenter about 16 km north of the city of Anchorage. Peak ground accelerations reached ∼30% g. Despite the relatively low severity of most of the ground failure occurrences, geotechnical damage to buildings and structures was widespread (Franke and others, 2019). Here, we present an inventory of the earthquake-triggered ground failure based on information compiled from numerous data sources. The inventory is comprised of 886 points that each correspond to the location of earthquake triggered ground failure associated with the 2018 event (Figure 1). For each instance of ground failure, the direct observation is reported and, where possible, is categorized based on the process interpreted to have caused it. For example, an observation of a rock fall would be categorized as a landslide. Similarly, a sand boil would be categorized as evidence of liquefaction. Where possible, the features were more precisely mapped with an additional polygon (n=179) and or line feature (n=32). To determine the existence and location of ground failure features, we relied on a variety of mapping methods and data sources. As such, the certainty in the mapping of each feature within the inventory varies (Figure 2). Details on how uncertainty is reported as well as the methods and data sources used are discussed in the following sections.

Data and Methods The inventory was compiled using a variety of different sources and methods with varying levels of success (Martinez and others, 2021). To convey this, the overall mapping certainty for each feature within the inventory is determined by assigning a semi-quantitative grade for categories related to the certainty of the existence of the feature, the positional accuracy of the mapped feature, and the mapped delineation quality. Those categories are denoted as presence, location, and delineation certainty in the inventory attribute table. Confidence grade definitions, and all other inventory attributes, are described in Table 1. As previously mentioned, we relied on a variety of methods and data sources to develop the ground failure inventory. We primarily relied on field observation data and information shared from federal, state, and local agencies. These data sources were particularly helpful where remotely sensed data were not available or insufficient for mapping ground failure. Here, we summarize how remotely sensed data were used and in which ways they were advantageous or insufficient. High resolution digital elevation models (DEMs) are typically used to develop high-quality ground failure inventories. We began our ground failure mapping efforts by differencing high resolution pre- and post-earthquake lidar derived DEMs from the Alaska Department of Natural Resources (AK DNR) Division of Geological and Geophysical Surveys (DGGS) elevation portal (DGGS Elevation Portal (alaska.gov)). The pre- and post-earthquake DEMs used to identify and map the extent of ground failure in this study had an areal overlap of approximately 321 km2. These data cover only a small portion of the area that experienced moderate to severe shaking during the earthquake (60,690 km2) as estimated from the USGS ShakeMap.

In regions lacking sufficient DEM coverage we used Normalized Difference Vegetation Index (NDVI) differencing to identify and map ground failure. NDVI maps are derived from multispectral imagery and display the relative health or existence of vegetation in a landscape. For this study, NDVI maps were derived using imagery from the European Space Agency’s Sentinel-2 (S2) Multispectral Instrument (MSI) (Drusch and others, 2012). Ground failure that is severe enough to disturb vegetation can be mapped by differencing pre- and post- earthquake NDVI maps. However, efforts to create NDVI maps of the landscape following the earthquake were hindered due to snowfall that began to accumulate shortly after the earthquake (2 December 2018). To overcome this limitation, we used Google Earth Engine (Gorelick and others, 2017) to create composites of multispectral images of the summers (May 1 to July 30) before and after the earthquake instead. The image composites were then used to create NDVI maps that were differenced to determine where there were any significant changes in vegetation that could potentially correspond to ground failure. While the resolution of the data used for the NDVI differencing (10-m) is lower than that used for DEM differencing (1-m) it is advantageous as it is available globally, and thus, for the entire earthquake affected region of Southcentral, Alaska.

Despite the advantage of globally available NDVI data, the method is still limited in its ability to detect smaller and more subtle vegetation changes. Our comparisons may also include some landslides that were not triggered by the earthquake but happened in the time spanned by the images. When possible, satellite and aerial imagery alone were also used to map ground failure features. However, in most instances the imagery alone was inadequate for mapping because many features were not visible in available satellite imagery. In addition, the ground surface was obscured in satellite imagery as a result of snow fall that occurred shortly after the earthquake. Some features, such as large road cracks, were not visible following snowmelt as they were quickly repaired by The Alaska Department of Transportation and Public Facilities. Other ephemeral features, such as sand boils or cracks on tidal flats, were no longer visible in satellite imagery after snowmelt and several tidal cycles. Remotely sensed data alone were found to be insufficient for mapping ground failure associated with this event (see Martinez and others, 2021). In addition to the limitations of remotely sensed data, the aforementioned snowfall that occurred immediately after the earthquake obscured large swaths of land and subsequently hindered efforts to gather ground failure observations while in the field. Thus, in addition to the remotely sensed data, the inventory was supplemented with field observations gathered by U.S. Geological Survey scientists shortly after the event (Grant and others, 2020), information on earthquake damage supplied from federal (FEMA Region X), state (The Alaska Department of Transportation and Public Facilities) and local agencies (Municipality of Anchorage, Kenai Peninsula Borough, Matanuska-Susitna Borough Department of Emergency Services, 673 Civil Engineers Squadron at the Joint Base Elmendorf-Richardson, Anchorage Fire Station 12) as well as compiled social media information and data from literature on the 2018 earthquake event. Using the data mentioned above, we inferred which ground failure process to assign to each ground failure location based on contextual information such as the environment and presence of evidence that suggests ground failure may have occurred. We assigned the “landslide” process to locations where there are indications of the downhill movement of material. A modified classification scheme based on Keefer (1984) was used to classify the observed landslide features within our inventory (see Table 1 for landslide classification scheme). Evidence that liquefaction processes may have occurred include the presence of ground cracks, sand boils, lateral spreads, the settlement of structures, and rapid soil flows. In cases where direct evidence of liquefaction is lacking (i.e., sand boils), the classification is based on proximity to identified sand boils. Liquefaction was interpreted to be the process for observed lateral spreading features parallel to water bodies. Low-lying features on flat slopes (e.g., settlement alone) for which there is no direct evidence to suggest that liquefaction occurred are given the label “Liquefaction (Ambiguous)” to reflect the uncertain nature in their process designation.
Despite the use of supplemental data, the inventory is still considered incomplete, with varying levels of completeness throughout the region. Adverse environmental conditions including snow fall, limited daylight hours, a large earthquake-affected area, and accessibility constraints limited the extent to which both field and remotely sensed observations could be reliably used to compile a ground failure inventory. To communicate this variability in inventory completeness and quality, Figure 2 displays the region, highlighted in white, in which direct observations of ground failure were made in addition to where high-resolution pre- and post-earthquake lidar data are available. This higher-confidence mapping region displays where the inventory is considered complete and high quality due to the nature of the data (i.e., direct observations, high-quality lidar) used for mapping in this region. To minimize human error, two co-authors carefully reviewed the inventory for inconsistencies and errors.

Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government. References Cited: Drusch, M., Del Bello, U., Carlier, S., Colin, O., Fernandez, V., Gascon, F., Hoersch, B., Isola, C., Laberinti, P., Martimort, P., Meygret, A., Spoto, F., Sy, O., Marchese, F., and Bargellini, P., 2012, Sentinel-2: ESA’s Optical High-Resolution Mission for GMES Operational Services: Remote Sensing of Environment, v. 120, p. 25–36. https://doi.org/10.1016/j.rse.2011.11.026 Franke, K.W., Koehler, R., Beyzaei, C.Z., Cabas, A., Christie, S., Dickenson, S., Pierce, I., Stuedlein, A., and Yang, Z., 2019, Geotechnical Engineering Reconnaissance of the 30 November 2018 Mw 7.1 Anchorage, Alaska Earthquake, Version 2.0: Geotechnical Extreme Events Reconnaissance Association. Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., and Moore, R., 2017, Google Earth Engine: Planetary-scale geospatial analysis for everyone: Remote Sensing of Environment, v. 202, p. 18–27. https://doi.org/10.1016/j.rse.2017.06.031 Grant, A.R., Jibson, R.W., Witter, R.C., Allstadt, K., Thompson, E., Bender, A.M and Schmitt, R.G., 2020, Field reconnaissance of ground failure triggered by shaking during the 2018 M7.1 Anchorage, Alaska, earthquake: U.S. Geological Survey data release, https://doi.org/10.5066/P99ONUNM Jibson, R.W., R. Grant, A.R., Witter, R.C., Allstadt, K.E., Thompson, E.M., and Bender, A.M., 2020, Ground Failure from the Anchorage, Alaska, Earthquake of 30 November 2018: Seismological Research Letters, v. 91, no. 1, p. 19–32. https://doi.org/10.1785/0220190187 Keefer, D.K., 1984, Landslides caused by earthquakes. Geological Society of America Bulletin, 95(4), pp.406-421. Martinez, S.N., Schaefer, L.N., Allstadt, K.E., and Thompson, E.M., 2021, Evaluation of Remote Mapping Techniques for Earthquake-Triggered Landslide Inventories in an Urban Subarctic Environment: A Case Study of the 2018 Anchorage, Alaska Earthquake: Frontiers in Earth Science, v. 9. https://doi.org/10.3389/feart.2021.673137

Access & Use Information

Public: This dataset is intended for public access and use. License: No license information was provided. If this work was prepared by an officer or employee of the United States government as part of that person's official duties it is considered a U.S. Government Work.

Downloads & Resources

Dates

Metadata Created Date June 1, 2023
Metadata Updated Date December 31, 2023

Metadata Source

Harvested from DOI EDI

Additional Metadata

Resource Type Dataset
Metadata Created Date June 1, 2023
Metadata Updated Date December 31, 2023
Publisher U.S. Geological Survey
Maintainer
@Id http://datainventory.doi.gov/id/dataset/f43ca78080bb250400c85ca3262316d1
Identifier USGS:61d49498d34ed79293fe9095
Data Last Modified 20231229
Category geospatial
Public Access Level public
Bureau Code 010:12
Metadata Context https://project-open-data.cio.gov/v1.1/schema/catalog.jsonld
Metadata Catalog ID https://datainventory.doi.gov/data.json
Schema Version https://project-open-data.cio.gov/v1.1/schema
Catalog Describedby https://project-open-data.cio.gov/v1.1/schema/catalog.json
Harvest Object Id a8d9f5fe-26f1-431f-89dc-64a25c70a4c4
Harvest Source Id 52bfcc16-6e15-478f-809a-b1bc76f1aeda
Harvest Source Title DOI EDI
Metadata Type geospatial
Old Spatial -151.8365,59.7949,-147.6825,62.2778
Publisher Hierarchy White House > U.S. Department of the Interior > U.S. Geological Survey
Source Datajson Identifier True
Source Hash 5f51da0c283b8fc9a6ed22cc7da54ac27fcd0d9812f8a22323311c3ab0cb3cd9
Source Schema Version 1.1
Spatial {"type": "Polygon", "coordinates": -151.8365, 59.7949, -151.8365, 62.2778, -147.6825, 62.2778, -147.6825, 59.7949, -151.8365, 59.7949}

Didn't find what you're looking for? Suggest a dataset here.