Holt, James
(2019)
Addressing Uncertainty in Earthquake Magnitudes Commonly Used in Modern Seismic Hazard Assessment.
PhD thesis, University of Liverpool.
|
Text
200880943_Sep2019.pdf - Unspecified Download (16MB) | Preview |
Abstract
Earthquake magnitudes are central to seismic hazard assessment filtering in as one of the primary characteristics of earthquakes to which we can develop spatio-temporal relationships as well as scaling relationships. As such it is critical to constrain earthquake magnitudes and report accurate uncertainty and find strategies to reduce it. To that end, we explore methods to either report accurate uncertainty or directly improve magnitude determination by using high quality seismic data from three regions (Utah, USA, Yellowstone National Park, USA and Continental Japan). Our first study is focussed on Utah, USA in cooperation with the University of Utah Seismograph stations (UUSS). Here we develop the capability to estimate the moment magnitude (Mw), commonly used in modern seismic hazard assessment (SHA) studies, applying a spectral method (Edwards et al., 2010) to small and moderate-sized earthquakes. Using our improved methodology, we produce a new Mw catalogue containing > 200 Mw estimates (1.5 < Mw < 4.0). Furthermore, with our expanded Mw dataset for the Utah region we find bilinear relationships for Mw as a function of both ML and coda-duration magnitude (Mc), thus providing an improvement to those derived previously for the region (Arabasz et al., 2016). The slope of the ML - Mw relationship is 2/3 for ML < 3, consistent with theory and other similar experimental studies. Our second study, also ran in cooperation with UUSS, is directed toward reducing uncertainty in local magnitude (ML) calculation, for earthquakes recorded in Yellowstone National Park, USA. A non-parametric inversion (Savage and Anderson, 1995) is used to derive a new distance correction for Yellowstone, along with a new set of station corrections. We find Yellowstone has a unique and complex local attenuation structure. Additionally, the new calculated ML, using our new distance correction and station corrections, provides improved agreement with the moment magnitude (Mw) estimates (3.1 < Mw < 4.5) than the previous scale. Furthermore, we find evidence that the old station corrections traded off with the old distance correction (Richter, 1958). In our final study we explore the role of scenario-dependent site amplification on local magnitude (ML) on a single-station basis. In order to isolate the effect of the near surface amplification on ML, relative differences between station-specific ML at the surface and borehole (ΔML,STN) are studied for 34 sites from the KiK-Net network, Japan. We find strong moment magnitude (M) dependent, scenario specific, ΔML,STN trends. To further our understanding, we developed stochastic models to simulate the response and understand this variability. Simulated data, ΔML,STN(M), based on the available site response information are shown to closely match the empirical ΔML,STN trends. We conclude by recommending how these models may be used as station corrections to reduce uncertainty and bias in the determination of ML.
| Item Type: | Thesis (PhD) |
|---|---|
| Divisions: | Faculty of Science and Engineering > School of Environmental Sciences |
| Depositing User: | Symplectic Admin |
| Date Deposited: | 08 Jan 2020 10:06 |
| Last Modified: | 19 Jan 2023 00:17 |
| DOI: | 10.17638/03064483 |
| Supervisors: |
|
| URI: | https://livrepository.liverpool.ac.uk/id/eprint/3064483 |
Dimensions
Dimensions
