Please use this identifier to cite or link to this item: https://www.um.edu.mt/library/oar/handle/123456789/79083
Title: Automatic earthquake detection and localisation from a three-component single-station
Authors: Agius, Matthew Richard (2007)
Keywords: Earthquakes
Seismology -- Malta
Seismology -- Technique
Issue Date: 2007
Citation: Agius, M. R. (2007). Automatic earthquake detection and localisation from a three-component single-station (Master’s dissertation).
Abstract: As more and more seismic stations are being deployed, seismic networks are enhancing their distribution and analysing earthquake location better. Usually networks detect an earthquake when a number of stations trigger and then analyse the different times-of-arrival to determine the source. With the installation of broadband seismometers, instruments are more sensitive to local seismic activity, but due to the large volume of data that load the network, and the threshold set to detect an event, data from single stations is being left unanalysed. The Physics Department of the University of Malta operates a single, 3-component broadband digital seismograph at Wied Dalam, l/o Birzebbugia in the south of Malta (WDD). Earthquakes in the Sicily Channel have always been difficult to locate, either because they fall outside the southern Italian network, leading to inaccurate location, or because they are poorly recorded on any station other than WDD, on Malta. This work has addressed the problem of single-station recording of earthquakes in the Sicily Channel by designing and implementing an automated system for detecting, identifying and locating such events, using 3-component single-station polarization analysis. The system, nicknamed LESSLA (Local Earthquake Single-Station Location Analyser) utilizes the three components from each of the three sampling streams HH, BH and LH from the Quanterra data acquisition systems, to identify, and pick, major arrivals such as P and S. The standard STA/LTA arrival detection algorithm has been used, and improved for better accuracy. For local and regional earthquakes, event distance is calculated from measured S-P times against a calibrated regional travel time graph, while event azimuth is measured using all three components, after the algorithm of Roberts, Christoffersson and Cassidy (1989). LESSLA uses a weighting scheme on the three different sampling streams to classify an event as a True event of local I regional origin. The system analyses one whole day of data and issues a detailed daily bulletin, containing pick times, event distance, azimuth and location, magnitude, epicentre map, PDF files of event seismograms, links to relevant international bulletins and other analysis information. The report is also sent by email, allowing the user to visually analyse the event seismograms and make rapid judgement on the location reliability The method has been implemented and succeeded to locate regional and local events while also identifying quarry blasts. 70% of the earthquakes under test had both P and S-pick correct and 70% of earthquakes had a 30° or less azimuth variation with respect to INGV I EMSC bulletins. Moreover, more than 17 events were located along a documented fault system to the south and south-east of Malta, which were not reported by any other network. The seismicity of this feature had not been previously identified.
Description: M.SC.PHYSICS
URI: https://www.um.edu.mt/library/oar/handle/123456789/79083
Appears in Collections:Dissertations - FacSci - 1965-2014
Dissertations - FacSciPhy - 1967-2017

Files in This Item:
File Description SizeFormat 
M.SC._Agius_Matthew Richard_2007.pdf
  Restricted Access
16.83 MBAdobe PDFView/Open Request a copy
Agius_Matthew_Richard_acc.material.pdf
  Restricted Access
215.42 kBAdobe PDFView/Open Request a copy


Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.