Preserving the World Heritage: Post-Earthquake Monitoring Based on Structural Break Testing with Deep Temporal Convolutional Features - First Torsional Mode dataset

The dataset consists of two primary data sources: i) Seismic data recorded by a triaxial seismometer installed in the south wall of the church's nave, at a height of around 6 meters, with a sampling rate of 100 Hz. The natural frequencies are extracted from this data using the Random Decrement Technique \cite{RDT_nasa,RDT_ardeche}; ii) Environmental data obtained from a Davis Vantage Pro 2 weather station located approximately 3.2 km from the church, providing measurements of temperature, humidity, wind speed, atmospheric pressure, and rainfall. Both time series are at a resolution of 1-hour. The corresponding paper “Preserving the World Heritage: Post‑Earthquake Monitoring Based on Structural Break Testing with Deep Temporal Convolutional Features.” was accepted at ECML-PKDD2025 (Applied Data Science track) to be held in Porto in September 2025. Authors: Francesco Dente, Andy Combey, Alix Lhéritier, Rodrigo Acuna-Agost, Diego Mercerat.