Bayesian Estimations of Peak Ground Acceleration and 5% Damped Spectral Acceleration from Modified Mercalli Intensity Data
John E. Ebel, David J. Wald
Abstract
John E. Ebel, David J. Wald
Abstract
We describe a new probabilistic method that uses observations of modified Mercalli intensity (MMI) from past earthquakes to make quantitative estimates of ground shaking parameters (i.e., peak ground acceleration, peak ground velocity, 5% damped spectral acceleration values, etc.). The method uses a Bayesian approach to make quantitative estimates of the probabilities of different levels of ground motions from intensity data given an earthquake of known location and magnitude. The method utilizes probability distributions from an intensity/ground motion data set along with a ground motion attenuation relation to estimate the ground motion from intensity. The ground motions with the highest probabilities are the ones most likely experienced at the site of the MMI observation. We test the method using MMI/ground motion data from California and published ground motion attenuation relations to estimate the ground motions for several earthquakes: 1999 Hector Mine, California (M7.1); 1988 Saguenay, Quebec (M5.9); and 1982 Gaza, New Hampshire (M4.4). In an example where the method is applied to a historic earthquake, we estimate that the peak ground accelerations associated with the 1727 (M∼5.2) earthquake at Newbury, Massachusetts, ranged from 0.23 g at Newbury to 0.06 g at Boston.
OpenAlex reports 13 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We describe a new probabilistic method that uses observations of modified Mercalli intensity (MMI) from past earthquakes to make quantitative estimates of ground shaking parameters (i.e., peak ground acceleration, peak ground velocity, 5% damped spectral acceleration values, etc.). The method uses a Bayesian approach to make quantitative estimates of the probabilities of different levels of ground motions from intensity data given an earthquake of known location and magnitude. The method utilizes probability distributions from an intensity/ground motion data set along with a ground motion attenuation relation to estimate the ground motion from intensity. The ground motions with the highest probabilities are the ones most likely experienced at the site of the MMI observation. We test the method using MMI/ground motion data from California and published ground motion attenuation relations to estimate the ground motions for several earthquakes: 1999 Hector Mine, California (M7.1); 1988 Saguenay, Quebec (M5.9); and 1982 Gaza, New Hampshire (M4.4). In an example where the method is applied to a historic earthquake, we estimate that the peak ground accelerations associated with the 1727 (M∼5.2) earthquake at Newbury, Massachusetts, ranged from 0.23 g at Newbury to 0.06 g at Boston.
Key concepts: Mercalli intensity scale, Peak ground acceleration, Spectral acceleration, Attenuation, Intensity (physics), Acceleration, Ground motion, Geodesy