DOGAMI Open-File Report Series

[Go back] | Publications Center | DOGAMI Home

Publication Preview

Open-File Report O-15-01, Landslide susceptibility analysis of lifeline routes in the Oregon Coast Range, by Rubini Mahalingam, Michael J. Olsen, Mahyar Sharifi-Mood, and Daniel T. Gillins, Oregon State University School of Civil and Construction Engineering.

Download .zip file (3.63 GB)

Note: This publication is one outcome of an Oregon Department of Transportation Research project. A second publication is ODOT Research Report SPR-740, Impacts of Potential Seismic Landslides on Lifeline Corridors, by Michael J. Olsen, Scott A. Ashford, Rubini Mahalingam, Mahyar Sharifi-Mood, Matt O’Banion, and Daniel T. Gillins, Oregon State University School of Civil and Construction Engineering. 

ODOT, OSU and DOGAMI have been state leaders in research on risks posed by the Cascadia subduction zone, possible earthquake and tsunami impacts, and initiatives to help the state prepare for a future disaster that is considered a certainty by scientists.

DOGAMI publication includes:

geodatabase icon GIS folder:

Geodatabase:
Esri v10.1 formatted Geodatabase. Metadata are embedded in the geodatabase.
Landslide_Susceptibility_Analysis_of_Lifeline_Routes_in_Oregon.gdb:
feature dataset classes:
- G_FAULT_LN_UTM (lines),
- G_MAP_UNIT_UTM (polygons),
rasters:
- Aspect
- Aspect_300m
- Dist_Fault_km
- Dist_River_Km
- NDVI
- ORP_PGA_M9
- ORP_PGV_M9
- pls_30m, pls_30m_hwy
- Precipitation_30m
- Slope_30m

Layer_Files:
- Generalized Geology.lyr
- LandslideDatabase.lyr
- NDVI.lyr
- PGA_M9.lyr
- PGV_M9.lyr
- PLS_30m.lyr
- PLS_30m_HWY.lyr
- PLS_30m_HWY.lyr.xml
- Precipitation.lyr
- Slope.lyr

KML folder:

- kml.kmz
- raster.kmz

MODELBUILDER_SCRIPT folder:

- model_builder_script


MAP_PDF folder:

PDF file format; dimensions 46 x 60 inches. Plate comes in two versions, one suitable for relatively small file size onscreen viewing and one suitable for high-resolution printing. A small (1008 x 774 pixel) browse graphic jpg and a 300-dpi tiff of the map plate are also included.

Counties or parital counties covered: Benton, Clackamas, Clatsop, Columbia, Coos, Curry, Douglas, Jackson, Josephine, Lane, Lincoln, Linn, Marion, Multnomah, Polk, Tillamook, Washington, Yamhill

Abstract from the map plate:

Coastal communities in Oregon are highly dependent on only a few lifeline highway and utility corridors that locally route across the Coast Range, connecting the coast with the Willamette Valley and the Interstate 5 lifeline corridor. However, these routes are frequently threatened by landslides, which can damage the roadway, block traffic, and threaten life safety. Understanding how landslides can impact these routes is important for planning, preparation, and resilience purposes. To this end, we analyzed the contribution of numerous factors to landslide triggering by using multiple linear regression analysis. In this analysis, factors including vector datasets such highway locations, faults, folds, geology/lithology, streams, and land use along with raster datasets such as slope, aspect, slope roughness, terrain (elevation) roughness, peak ground acceleration (PGA) and peak ground velocity (PGV) from a Cascadia scenario event, and precipitation were correlated with the locations of previously mapped landslides. A predictive equation to determine the likelihood of a landslide event spatially across the entire study area was created from the most influential parameters. Given the wide variety of parameters, data accuracies, and variability across the study area, we developed automated batch-processing tools to perform this regional analysis and generate these maps in a Geographic Information System (GIS) domain. The selected model was validated using a validity index, which compares the percentage area of occurrence of landslides from the inventory maps with the susceptibility map. To provide context to the potential impact of a landslide on a highway, the “fuzzy AND” technique was used to classify each pixel on the basis of its relative position to the highway and in the susceptibility map. The analysis also indicated that slope, PGA, PGV, and precipitation were the most statistically significant factors controlling landslide movements throughout the coastal region of Oregon. The influence of other factors was generally minimal in comparison.

O-15-01 map plate