TY  - JOUR
A1  - Lilienkamp, Henning
A1  - von Specht, Sebastian
A1  - Weatherill, Graeme
A1  - Caire, Giuseppe
A1  - Cotton, Fabrice
T1  - Ground-Motion modeling as an image processing task
BT  - introducing a neural network based, fully data-driven, and nonergodic
JF  - Bulletin of the Seismological Society of America
N2  - We construct and examine the prototype of a deep learning-based ground-motion model (GMM) that is both fully data driven and nonergodic. We formulate ground-motion modeling as an image processing task, in which a specific type of neural network, the U-Net, relates continuous, horizontal maps of earthquake predictive parameters to sparse observations of a ground-motion intensity measure (IM). The processing of map-shaped data allows the natural incorporation of absolute earthquake source and observation site coordinates, and is, therefore, well suited to include site-, source-, and path-specific amplification effects in a nonergodic GMM. Data-driven interpolation of the IM between observation points is an inherent feature of the U-Net and requires no a priori assumptions. We evaluate our model using both a synthetic dataset and a subset of observations from the KiK-net strong motion network in the Kanto basin in Japan. We find that the U-Net model is capable of learning the magnitude???distance scaling, as well as site-, source-, and path-specific amplification effects from a strong motion dataset. The interpolation scheme is evaluated using a fivefold cross validation and is found to provide on average unbiased predictions. The magnitude???distance scaling as well as the site amplification of response spectral acceleration at a period of 1 s obtained for the Kanto basin are comparable to previous regional studies.
Y1  - 2022
U6  - https://doi.org/10.1785/0120220008
SN  - 0037-1106
SN  - 1943-3573
VL  - 112
IS  - 3
SP  - 1565
EP  - 1582
PB  - Seismological Society of America
CY  - Albany
ER  -