Data and code for "Peak ground acceleration underestimates narrow-band seismic amplification on mountain slopes"

DATA AND CODE FOR:Peak ground acceleration underestimates narrow-band seismic amplification on mountain slopes OVERVIEW--------This repository contains the derived data products, documentation,and Python code supporting the analyses reported in the accompanyingmanuscript on seismic amplification at a mountain ridge in Totsukawa,Japan. The deposited archive includes an event catalogue, phase picks,peak-ground-acceleration data, per-event spectral-ratio peaks,full station-to-valley spectral-ratio curves, near- and far-eventmean spectral-ratio curves, back-azimuth and aspect-angle data,a Jupyter reproducibility notebook, a file manifest, validationresults, and source-provenance documentation. The notebook regenerates the reported numerical analyses andanalytical equivalents of selected data-derived figures, including: - PGA amplification statistics and distance relationships- Per-event spectral-ratio peak statistics- Near- and far-event mean spectral-ratio curves- Correlations with epicentral distance, focal depth, and the angle between back-azimuth and station aspect- Analytical equivalents of main-text Figs. 2–4 and Supplementary Fig. S3 Table 3 values are reproduced exactly. Small differences in theTable 1 correlation coefficients arise because the deposited PGAvalues are rounded to two decimal places in the supplementary table. DATA AVAILABILITY-----------------The repository contains processed and derived data products ratherthan raw accelerometer waveforms. Requests for access to the originalwaveform records may be directed to the corresponding author. CONTENTS--------The downloadable archive contains: - README.txt- reproduce_figures_tables.ipynb- repository_file_manifest.csv- source_provenance.md- validation_report.md- event catalogue and phase-pick tables- PGA amplification tables- spectral-ratio peak and full-curve data- near/far mean spectral-ratio data- back-azimuth and aspect-angle data SOFTWARE REQUIREMENTS---------------------Python 3.10 or later, with NumPy, SciPy, pandas, matplotlib,and Jupyter or nbconvert. LICENCE-------The deposited dataset is released under the Creative CommonsAttribution 4.0 International licence.