<!-- Our title --> <div align="center"> <h3 style="font-size: 25px;">hvsrprocpy</h3> </div> <!-- Short description --> <p align="center"> A python library that performs Horizontal-to-Vertical Spectral Ratio (HVSR) processing. </p>

<div align="center"> <h3 style=" font-size: 25px;">Authors</h3> </div> <!-- Short description --> <p align="center"> Francisco Javier G. Ornelas¹, Pengfei Wang², Scott J. Brandenberg¹, Jonathan P. Stewart¹ </p>

¹ University of California, Los Angeles (UCLA) <br> ² Old Dominion University<br>

<div style="text-align: center;"> <h3 style=" font-size: 25px;">Table of Contents</h3> </div>

• Introduction
• Background
• Getting started
• Citation
• Issues

Introduction

---

A python library that can perform Horizontal-to-Vertical Spectral Ratio (HVSR) processing from recordings of microtremors or earthquakes from 3-component seismometers. This library was developed by Francisco Javier G. Ornelas under the supervision of Dr. Jonathan P. Stewart and Dr. Scott J. Brandenberg at the University of California, Los Angeles (UCLA). Other contributions came from Dr. Pengfei Wang a professor at Old Dominion University, who wrote hvsrProc a rstudio package, which this python library is based on. That work can be found here:

Wang, P. wltcwpf/hvsrProc: First release (Version v1.0.0). Zenodo. http://doi.org/10.5281/zenodo.4724141

Background

---

HVSR is derived from ratios of the horizontal and vertical components of a Fourier Amplitude Spectrum (FAS) from a 3-component recording of microtremors or earthquakes. This is done by recording ground vibrations either from temporarily-deployed or permanently-installed seismometers, for a relatively short period of time (~1-2 hrs) or a longer period of time.

This method or technique was first proposed by Nogoshi and Igarashi (1971) <a href="https://www.scirp.org/reference/referencespapers?referenceid=3100696" target="_blank">[ABSTRACT]</a> and later popularized by Nakamura (1989) <a href="https://trid.trb.org/View/294184" target="_blank">[ABSTRACT]</a>. The method works by assuming that the horizontal wavefield is amplified as seismic waves propagate through the soil deposits compared to the vertical wavefield.

HVSR can be useful in site characterization since it can identify resonant frequencies at sites, through peaks in an HVSR spectrum. Studies have also found that the lowest peak in HVSR spectra can be associated with the fundamental frequency of a site (e.g., <a href="https://link.springer.com/article/10.1007/s10518-012-9413-4" target="_blank">[ABSTRACT]</a>).

Getting started

---

Installation

hvsrprocpy is available using pip and can be installed with:

• Jupyter Notebook pip install hvsrprocpy
• PyPI py -m pip install hvsrprocpy

Usage

hvsrprocpy is a library that performs hvsr related processing. The library contains various features, such as:

• Manual selection of windows in the time domain and the frequency domain.
• Rotated HVSR to see the azimuthal variability of HVSR.
• Different distibutions such as normal or log-normal.
• Different smoothing functions such as Konno and Ohmachi and Parzen smoothing.
• Different type of horizontal combinations, such as geometric mean, squared average, and RotD50.
• Various outputs such as mean FAS and HVSR, selected HVSR and ts, and rotated HVSR.

Examples of these can be found under the examples folder in the Github repository <a href="https://github.com/fjornelas/hvsrprocpy" target="_blank">[GIT]</a>

Example of manual window selection on time series

<img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/microtremor_ts_win_sel_example.png?raw=true" width="775">

Example of manual window selection on HVSR

<img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/hvsr_fas_win_sel_example.png?raw=true" width="775">

Example of azimuthal plots

<img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/hvsr_polar_example.png?raw=true" width="775">

Example of Mean Curve plot with Metadata

<img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/hvsr_mean_curve_and_meta_example.png?raw=true" width="775">

Example comparisons of Konno and Ohmachi (left) and Parzen smoothing (right) on FAS

<div style="display: flex; flex-direction: row;"> <img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/fas_ko_smoothing_example.png?raw=true" width="400" style="margin-right: 10px;" title="FAS KO Smoothing Example"> <img src="https://github.com/fjornelas/hvsrprocpy/blob/main/figs/fas_parzen_smoothing_example.png?raw=true" width="400" title="FAS Parzen Smoothing Example"> </div>

Citation

---

If you use hvsrprocpy (directly or as a dependency of another package) for work resulting in an academic publication or other instances, we would appreciate if you cite the following:

Ornelas, F. J. G., Wang, P., Brandenberg, S. J., & Stewart, J. P. (2024). hvsrprocpy (Version 1.1.0) [Computer software]. https://doi.org/10.5281/zenodo.13352470

Issues

---

Please report any issues or leave comments on the <a href="https://github.com/fjornelas/hvsrprocpy/issues" target="_blank">Issues</a> page.

License

This project has been licensed under more information about the license can be found here <a href="https://github.com/fjornelas/hvsrprocpy/blob/main/LICENSE" target="_blank">[LICENSE]</a>.

Acknowledgements

We would also like to thank the many others who aided in the development of this python library, these are:

• John Stapleton
• Chukwuma Okonkwo
• Chukwuebuka C. Nweke
• Tristan Buckreis
• Christopher de la Torre

for their support in helping develop this python library.