Kushal Sharma Wagle, Adrian Rodriguez-Marek, Joseph P. Vantassel jpvantassel.com
mHVSR-Vs30
is a collection of data-driven models to predict the
time averaged shear wave velocity in the upper 30 m (Vs30), from
microtremor horizontal-to-vertical spectral ratio (mHVSR). The developed
models developed include both low-dimensional (low_dim_models.ipynb
) and
high-dimensional (high_dim_models.ipynb
). The details of the model's
development and performance are presented in the reference below.
If you use these tools in your research or consulting, we ask you please cite the following:
Sharma Wagle, K., Vantassel, J.P., and Rodriguez-Marek, A. (2025). "A Set of Data-Driven Models to Predict VS30 from the Horizontal-to-Vertical Spectral Ratio of Microtremors". Bulletin of the Seismological Society of America. [In-Review]
-
If you do not have Python 3.12 or later installed, you will need to do so. A detailed set of instructions can be found here.
-
MacOS users, note that you may need to install
libomp
before continuing. You can do this usingbrew
with the commandbrew install libomp
or throughconda
with the commandconda install -c conda-forge libomp
. -
Download a copy of the repository to your local machine. This can be done by downloading a .zip file or by cloning the repository. If you download a .zip you will need to be sure that you continue using the unzipped version to prevent issues.
-
Install the Python dependencies using
pip
via the commandpython -m pip install -r requirements.txt
. If you are not familiar withpip
, a useful tutorial can be found here. -
Confirm that the dependencies have installed/updated successfully by examining the last few lines of the text displayed in the console.
-
Launch the provided Jupyter notebooks,
low_dim_models.ipynb
orhigh_dim_models.ipynb
(recommended), for a no-coding-required introduction to prediction models. If you have not installedJupyter Lab
, detailed instructions can be found here. -
Continue to explore the three datasets provided by changing the file names and associated metadata in the notebooks. Once you feel comfortable running the example data you can then apply the models on any mHVSR data of interest.
-
Enjoy!