Dr Benyamin Motevalli1, Dr Robert Pickle2, Dr Pavel Golodoniuc1, Mrs. Neda Taherifar1, Mr Vincent Fazio3, Mr Yunlong Li1
1CSIRO Mineral Resources, 26 Dick Perry Ave, Kensington, Australia, 2Australian National University, Research School of Earth Sciences, Australian National University, Building 142 Mills Road, Acton, Australia, 3CSIRO Mineral Resources, Research Way, Clayton, Australia
Biography:
Dr Ben Motevalli is a Senior Software Engineer specialising in computational nanomaterials, data science, cloud solution architecture, and application development. Passionate about transforming exceptional scientific and engineering ideas into highly interactive applications, Ben has contributed significantly across various sectors. After completing his PhD, he joined the Applied Machine Learning team at Data61, focusing on machine learning solutions for material science. He later transitioned to the Oil & Gas and Energy sectors, where he specialized in codifying physical models into core computational engines, developing predictive models, automating engineering analysis processes, and delivering user-friendly interactive applications. Ben was instrumental in Worley's Digital team, focusing on delivering Consultancy-as-a-Service. Currently at CSIRO, his primary role involves aiding researchers in transforming their ideas and computational models into modern, scalable applications.
Abstract:
Seismology is a data-intensive field, and recent trends have amplified the volume and complexity of data acquisition. The growing use of ambient noise studies and machine learning-based phase pickers has shifted the focus from traditional event-based analysis to continuous data streams. This shift has increased demands on researchers to develop scripting and data management skills simply to access and prepare seismic data. These technical barriers may deter new researchers and hinder broader adoption of the miniSEED format and FDSN (Federation of Digital Seismograph Networks) web services standard.
To address these challenges, we developed SEED-Vault—a cross-platform, open-source seismic data management tool built in Python. SEED-Vault integrates a user-friendly web interface powered by Streamlit with a flexible command-line interface, making it accessible for both novice and advanced users. It leverages the ObsPy library to ensure strict adherence to miniSEED and FDSNWS dataselect, station, and event specifications, maximizing interoperability with global FDSN data services.
SEED-Vault supports three main workflows: (1) event-based searches followed by station selection, (2) station-based searches followed by relevant event retrieval, and (3) continuous data download based on station and time range. Users can search and retrieve data from multiple data centers, including those requiring authentication. By streamlining the seismic data access process, SEED-Vault reduces entry barriers, enhances reproducibility, and promotes best practices in data standardization.
This tool aims to support a broad spectrum of seismological research and training, contributing to the long-term sustainability and accessibility of seismic data practices.