International Journal of
Physical Sciences

  • Abbreviation: Int. J. Phys. Sci.
  • Language: English
  • ISSN: 1992-1950
  • DOI: 10.5897/IJPS
  • Start Year: 2006
  • Published Articles: 2568

Full Length Research Paper

Spectral discrimination between mining blasts and natural earthquakes: Application to the vicinity of Tunçbilek mining area, Western Turkey

Kekovalı K*, Kalafat D and Deniz P
National Earthquake Monitoring Center, Kandilli Observatory and Earthquake Research Institute, BoÄŸaziçi University 34684 Çengelköy, Ä°stanbul, Turkey.
Email: [email protected]

  •  Accepted: 11 June 2012
  •  Published: 13 September 2012

Abstract

 

The objective of our paper is to develop time and frequency techniques to characterize mine or quarry blasts and discriminate these events from natural earthquakes. We are examining well documented mine blasts in the Kütahya Province in western Turkey that is an ideal case study since there are frequent confirmed routine mining blasts at Tunçbilek mining area. Particularly, Turkey’s richest lignite reserve was detected with the statistical analysis include Tunçbilek region that has been undertaken by Kekovali et al. (2011). In this study, we used time and frequency domain analysis (S/P wave amplitude peak ratio, complexity, spectral ratio) of a set of known earthquake and mining blasts seismograms of 520 seismic events (2.3 ≤ Md ≤3.0) from Kandilli Observatory Earthquake Research Institute and National Earthquake Monitoring Center (KOERI-NEMC) seismic catalog between 2009 to 2011. Out of a total 520 records, 344 are related to probable mining blasts and 176 to earthquakes. However, a new approach estimated on Pe (power of event) value comprising of three variables (S/P wave amplitude peak ratio, complexity, spectral ratio) that can be suitable identification of quarry-mining blasts in the the seismic catalogs, is presented and discussed for Tunçbilek region in the study. In the discrimination analysis between the earthquakes and mining blasts showed that Pe analysis comprising of three variables (S/P wave amplitude peak ratio, complexity, spectral ratio) was used together to increase the reliability instead of the S/P wave amplitude peak ratio or complexity and spectral ratio in time-frequency domain. The classsification was obtained with acceptably high results of 99.6% using Pe analysis. For a future study, we suggested that this algorithm could be applied to different earthquake-mining regions that was developed independently in this study. Because of the presence of quarry-mining blasts among the earthquakes recorded by the KOERI-NEMC seismic network, the decontamination of the catalog is essential for a reliable seismic hazard assessment and understanding the seismogenic processes of Turkey.

 

Key words: Earthquake, mining blast, amplitude discrimination, time-frequency domain.