Moroccan Seismic Activity Dataset for Deep Learning Analysis (1909-2024)
Description
This dataset presents historical earthquake occurrence data for the Morocco region, covering the period from 1909 to 2024. It includes detailed seismic attributes and geospatial parameters compiled from reliable seismological database, intended for use in scientific research, especially in the field of deep learning-based earthquake prediction and early warning systems. The data is curated from the official source, United States geological Survey. The dataset is structured in CSV format and contains the following features such as date, time, latitude, depth, mag (Richter Scale), magnitude type (magType), Number of Stations (nst), gap, minimum distance to station (dmin), RMS, Network Code (net), Event Id (id), Last Updated Timestamp (updated), Location (place), type, horizontalError, depthError, magError, Number of Stations for Magnitude (magnst), status, locationSource and magSource . This dataset has been used for training and evaluating various deep learning algorithms such as Feedforward Neural Network (FNN) , Bi-directional LSTM and Long Short - Term Memory (LSTM) models, in order to identify spatiotemporal patterns and anomalies for predictive modeling of seismic activities and analyzing the performance of the models. File Format: .csv Region: Morocco Time period: 1909-2024 Target Users: Researchers, data scientists, disaster management professionals and geophysicists.
Files
Steps to reproduce
1. Visit the USGS earthquake catalog 2. Set the date range 3. Under Search region, select bounding box with: - Min latitude and Max latitude -Min longitude and Max longitude 4. Set output format to CSV 5. run the search and download the dataset 6. Perform the data cleaning and preprocessing as required 7. Use the dataset for deep learning model training and analysis
Institutions
- Amrita Vishwa Vidyapeetham