HR: 1340h
AN: S23A-0228    [Abstracts]
TI: Automated Regolith Thickness Estimation for Site Classification
AU: * Potter, A K
EM: anna.potter@ga.gov.au
AF: Geoscience Australia, GPO Box 378, Canberra, ACT 2601 Australia
AB: In assigning site classes for seismic hazard assessment, it is preferable to have a detailed understanding of the distribution of unconsolidated sediments and in situ weathered bedrock (i.e. regolith) across an area. At present, regolith thickness data is generally acquired from bore hole drilling records and pre-existing seismic data. This process is time consuming and, due to variable data coverage and quality, often provide results containing high uncertainty. Gallant and Dowling (2003) observe that in some environments a direct relationship exists between thickness of accumulated regolith and terrain attributes associated with valley bottoms. Their automated Multi-resolution Valley Bottom Flatness (MrVBF) index, which classifies landscapes according to how strongly they resemble a valley bottom based on terrain attributes. In the present study, the ability of the algorithm to predict regolith thickness is tested in the Sydney region, Australia. The index is found to be highly sensitive to its assumptions regarding changes in geomorphology, and alone does not predict regolith thickness at spatial resolutions useful for local scale site classification. MrVBF does, however, achieve a degree of success at broader spatial resolutions, suggesting that automated algorithms of this type may have potential application in seismic hazard assessment and other forms of geoscientific investigation. Gallant, J.C. and Dowling, T.I. (2003). A multi-resolution index of valley bottom flatness for mapping depositional areas. Water Resources Research, 39: 1347-1360.
DE: 7212 Earthquake ground motions and engineering seismology
DE: 7299 General or miscellaneous
DE: 9330 Australia
SC: Seismology [S]
MN: Fall Meeting 2005